[PROPOSAL] Livy Proposal for Apache Incubator

classic Classic list List threaded Threaded
12 messages Options
Reply | Threaded
Open this post in threaded view
|

[PROPOSAL] Livy Proposal for Apache Incubator

Sean Busbey-2
Dear Apache Incubator Community,

I'm excited to present for discussion a proposal to move Livy into
incubation. Livy is web service that exposes a REST interface for managing
long running Apache Spark contexts in your cluster. With Livy, new
applications can be built on top of Apache Spark that require fine grained
interaction with many Spark contexts.

The proposal is on the wiki at the following page as well as copied in the
email below:

https://wiki.apache.org/incubator/LivyProposal

In addition to welcoming feedback on the proposal, we are actively seeking
one or more additional mentors. We also have included a section for
interested folks to ensure they get added to the mailing lists, presuming
Livy gets accepted for incubation.

---- LivyProposal

= Abstract =

Livy is web service that exposes a REST interface for managing
long running Apache Spark contexts in your cluster. With Livy, new
applications can be built on top of Apache Spark that require fine grained
interaction with many Spark contexts.  

= Proposal =

Livy is an open-source REST service for Apache Spark. Livy
enables applications to submit Spark applications and retrieve results
without a co-location requirement on the Spark cluster.

We propose to contribute the Livy codebase and associated artifacts (e.g.
documentation, web-site context etc) to the Apache Software Foundation.

= Background =

Apache Spark is a fast and general purpose distributed
compute engine, with a versatile API. It enables processing of large
quantities of static data distributed over a cluster of machines, as well as
processing of continuous streams of data. It is the preferred distributed
data processing engine for data engineering, stream processing and data
science workloads. Each Spark application uses a construct called the
SparkContext, which is the application’s connection or entry point to the
Spark engine. Each Spark application will have its own SparkContext.

Livy enables clients to interact with one or more Spark sessions through the
Livy Server, which acts as a proxy layer. Livy Clients have fine grained
control over the lifecycle of the Spark sessions, as well as the ability to
submit jobs and retrieve results, all over HTTP.  Clients have two modes of
interaction: RPC Client API, available in Java and Python, which allows
results to be retrieved as Java or Python objects. The serialization and
deserialization of the results is handled by the Livy framework.  HTTP based
API that allows submission of code snippets, and retrieval of the results in
different formats.

Multi-tenant resource allocation and security: Livy enables multiple
independent Spark sessions to be managed simultaneously. Multiple clients
can also interact simultaneously with the same Spark session and share the
resources of that Spark session. Livy can also enforce secure, authenticated
communication between the clients and their respective Spark sessions.

More information on Livy can be found at the existing open source website:
http://livy.io/

= Rationale =

Users want to use Spark’s powerful processing engine and API
as the data processing backend for interactive applications. However, the
job submission and application interaction mechanisms built into Apache
Spark are insufficient and cumbersome for multi-user interactive
applications.

The primary mechanism for applications to submit Spark jobs is via
spark-submit
(http://spark.apache.org/docs/latest/submitting-applications.html), which is
available as a command line tool as well as a programmatic API. However,
spark-submit has the following limitations that make it difficult to build
interactive applications: It is slow: each invocation of spark-submit
involves a setup phase where cluster resources are acquired, new processes
are forked, etc. This setup phase runs for many seconds, or even minutes,
and hence is too slow for interactive applications.  It is cumbersome and
lacks flexibility: application code and dependencies have to be pre-compiled
and submitted as jars, and can not be submitted interactively.

Apache Spark comes with an ODBC/JDBC server, which can be used to submit SQL
queries to Spark. However, this solution is limited to SQL and does not
allow the client to leverage the rest of the Spark API, such as RDDs, MLlib
and Streaming.

A third way of using Spark is via its command-line shell, which allows the
interactive submission of snippets of Spark code. However, the shell entails
running Spark code on the client machine and hence is not a viable mechanism
for remote clients to submit Spark jobs.

Livy solves the limitations of the above three mechanisms, and provides the
full Spark API as a multi-tenant service to remote clients.

Since the open source release of Livy in late 2015, we have seen tremendous
interest among a diverse set of application developers and ISVs that want to
build applications with Apache Spark. To make Livy a robust and flexible
solution that will enable a broad and growing set of applications, it is
important to grow a large and varied community of contributors.

= Initial Goals =

Move existing codebase, website, documentation and mailing
lists to Apache-hosted infrastructure Work with the infrastructure team to
implement and approve our code review, build, and testing workflows in the
context of the ASF Incremental development and releases per Apache
guidelines

= Current Status =

The Livy project began at Cloudera, as a part of the Hue
project. Cloudera soon realized the broad applicability of Livy, and
separated it out into an independent project in Nov 2015.

== Releases ==

Livy has undergone two public releases, tagged here:

* https://github.com/cloudera/livy/releases/tag/v0.2.0
* https://github.com/cloudera/livy/releases/tag/v0.3.0

Tarballs and zip files were created for each release and hosted on github.
Upon joining the incubator, we will adopt a more typical ASF release
process.

== Source ==

Livy’s source is currently hosted on Github at:

https://github.com/cloudera/livy

This repository will be transitioned to Apache’s git hosting during
incubation.

== Code review ==

Livy’s code reviews are currently public and hosted on
github as pull request reviews at: https://github.com/cloudera/livy/pulls
The Livy developer community so far is happy with github pull request
reviews and hopes to continue this after being admitted to the ASF.

== Issue Tracking ==

Livy’s bug and feature tracking is hosted on JIRA at:
https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
contains bugs and development discussion dating back 1 year and will provide
an initial seed for the ASF JIRA

== Community Discussion ==

Livy has several public discussion forums:

* https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
* https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user

== Development Practices ==

The Livy project follows a review before commit philosophy. Every commit
automatically runs through the unit tests and generates coverage reports
presented as a pull request comment. Our experience with this process leads
us to believe that it helps ease new contributors into the project. They get
feedback quickly on common mistakes, lowering the burden on reviewers. Those
same reviewers get to lead by example, showing the new contributors that we
value feedback within our community even when changes are done by more
experienced folks.

== Meritocracy ==

We believe strongly in meritocracy when electing committers and PMC members.
In the past few months, the project has added two new committers from two
different organisations, in recognition of their significant contributions
to the project. We will encourage contributions and participation of all
types, and ensure that contributors are appropriately recognized.

== Community ==

Though Livy is relatively new as a standalone open source project, it has
already seen promising growth in its community across several organizations:
Cloudera is the original development sponsor for Livy Microsoft pushed the
development of the interpreter fixing high availability issues and adding
additional features.  Hortonworks has contributed the security features to
Livy allowing kerberos and impersonation to work with Spark IBM is starting
to make contributions to the Livy project A number of other patches
contributed by community members

Livy currently relies on Google Groups for mailing lists. These lists have
been active since the end of 2015/start of 2016. Currently, Livy’s user
mailing list has 173 subscribers and has hosted a total of 227 topic
threads. Livy’s developer list has 49 subscribers and has hosted 79 topic
threads.

== Core Developers ==

The early contributions to Livy were made by Cloudera engineers. In 2016,
engineers from Microsoft and Hortonworks joined the core developer
community.

== Alignment ==

Livy is built upon Apache Spark, and other Apache projects like Apache
Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
community connections combined with our focus on development practices that
emphasize community engagement with a path to meritocratic recognition
naturally align us with the ASF.

= Known Risks =
== Orphaned Products ==

The risk of Livy being abandoned is low because it is supported by three
major big-data software vendors.  Moreover, Livy is already used to power
multiple releases of services and products used in production.

== Inexperience with Open Source ==

Several of the initial committers are experienced open source developers,
several being committers and/or PMC members on other ASF projects (Spark,
YARN).

== Homogenous Developers ==

The project already has a diverse developer base. It has contributions from
3 major organisations (Cloudera, Microsoft and Hortonworks), and is used in
diverse applications, in diverse settings (On-Prem and Cloud).

== Reliance on salaried Developers ==

The existing contributors to the Livy project have been made by salaried
engineers from Cloudera, Microsoft and Hortonworks. Since there are three
major organisations involved, the risk of reliance on a single group of
salaried developers is mitigated. The Livy user base is diverse, with users
from across the globe, including users from academic settings. We aim to
further diversify the Livy user and contributor base.

== Relationships with other Apache projects ==

Livy is closely tied to the Apache Spark project and currently addresses the
scenarios for a REST based batch and interactive gateway for Spark jobs on
YARN. Given the growing number of integrations with Livy, keeping it outside
of Apache Spark aligns with the desire of the Apache Spark community to
reduce the number of external dependencies in the Spark project.
Specifically, the Apache Spark community has previously expressed a desire
to keep job servers independent from the project.<<FootNote(See, for
example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
Furthermore, while Livy common usage is closely tied to Spark deployments
right now, its core building blocks can be reused elsewhere.  Livy’s Remote
REPL could be used as a library for interactive scenarios in non-Spark
projects. In the future, integrations with cluster managers like Apache
Mesos and others could also be added.

The features provided by Livy have already been integrated with existing
projects like Jupyter and Apache Zeppelin for their interactive Spark use
cases. This validates the need for a project like Livy and provides an
active downstream user base that the Livy community can interact with to
seed future interest in the project.

Livy serves a similar purpose to Apache Toree (incubating) but differs in
making session management, security and impersonation a focal design point.

== An Excessive Fascination with the Apache Brand ==

The primary motivation for submitting Livy to the ASF is to grow a diverse
and strong community. We wish to encourage diverse organisations, including
ISVs, to adopt Livy and contribute to Livy without any concerns about
ownership or licensing.

= Documentation =

Documentation can be found on the Livy website http://livy.io/ The Livy web
site is version controlled on the ‘gh-pages’ branch of the above repository
Additional documentation is provided on the github wiki:
https://github.com/cloudera/livy/wiki APis are documented within the source
code as JavaDoc style documentation comments.

= Initial Source =

The initial source code for Livy is hosted at

https://github.com/cloudera/livy 

= Source and Intellectual Property submission plan =

The Livy codebase and web site is currently hosted on GitHub and will be
transitioned to the ASF repositories during incubation. Livy is already
licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
CCLAs from all committers.  There are, however, some contributions recently
from authors that have not signed the CCLA and ICLA. If necessary for a
successful SGA, we’ll seek the necessary documentation or replace the
contributions.

The “Livy” name is not a registered trademark. We will need to do a
trademark search and make sure it is available for the Apache Foundation
prior to graduation.

Cloudera currently owns the domain name: http://livy.io/ which will be
transferred to the ASF and redirected to the official page during
incubation.

= External Dependencies =

The list below covers the non-Apache dependencies of the project and their
licenses.

 * Jetty: Apache 2.0
 * Dropwizard Metrics: Apache 2.0
 * FasterXML Jackson: Apache 2.0
 * Netty: Apache 2.0
 * Scala: BSD
 * Py4J: BSD
 * Scalatra: BSD

Build/test-only dependencies:

 * Mockito: MIT
 * JUnit: Eclipse

= Required Resources =
== Mailing Lists ==

 * [hidden email] (PPMC)
 * [hidden email] (dev mailing list)
 * [hidden email] (User questions)
 * [hidden email] (subscribers shouldn’t be able to post)
 * [hidden email] (subscribers shouldn’t be able to post)

== Git Repository ==

git://git.apache.org/livy

== Issue Tracking ==

We would like to import our current JIRA project into the ASF JIRA, such
that our historical commit message and code comments continue to reference
the appropriate bug numbers.

= Initial Committers =

 * Marcelo Vanzin ([hidden email])
 * Alex Man ([hidden email])
 * Jeff Zhang ([hidden email])
 * Saisai Shao ([hidden email])
 * Kostas Sakellis ([hidden email])

= Affiliations =

The initial set of committers includes people employed by Cloudera and
Hortonworks as well as one person currently unaffiliated with an employer.

= Additional Interested Contributors =

Those interested in getting involved with the project as we enter incubation
are encourage to list themselves here.

 * < add here >

= Sponsors =
== Champion ==

 * Sean Busbey ([hidden email])

== Nominated Mentors ==

 * Bikas Saha ([hidden email])
 * Brock Noland ([hidden email])

== Sponsoring Entity ==

We ask that the Incubator PMC sponsor this proposal.


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

William GUO
+1

Griffin needs Livy to access Spark context.


Thanks,
William

On 5/20/17, 7:45 AM, "Sean Busbey" <[hidden email]> wrote:

    Dear Apache Incubator Community,
   
    I'm excited to present for discussion a proposal to move Livy into
    incubation. Livy is web service that exposes a REST interface for managing
    long running Apache Spark contexts in your cluster. With Livy, new
    applications can be built on top of Apache Spark that require fine grained
    interaction with many Spark contexts.
   
    The proposal is on the wiki at the following page as well as copied in the
    email below:
   
    https://wiki.apache.org/incubator/LivyProposal
   
    In addition to welcoming feedback on the proposal, we are actively seeking
    one or more additional mentors. We also have included a section for
    interested folks to ensure they get added to the mailing lists, presuming
    Livy gets accepted for incubation.
   
    ---- LivyProposal
   
    = Abstract =
   
    Livy is web service that exposes a REST interface for managing
    long running Apache Spark contexts in your cluster. With Livy, new
    applications can be built on top of Apache Spark that require fine grained
    interaction with many Spark contexts.  
   
    = Proposal =
   
    Livy is an open-source REST service for Apache Spark. Livy
    enables applications to submit Spark applications and retrieve results
    without a co-location requirement on the Spark cluster.
   
    We propose to contribute the Livy codebase and associated artifacts (e.g.
    documentation, web-site context etc) to the Apache Software Foundation.
   
    = Background =
   
    Apache Spark is a fast and general purpose distributed
    compute engine, with a versatile API. It enables processing of large
    quantities of static data distributed over a cluster of machines, as well as
    processing of continuous streams of data. It is the preferred distributed
    data processing engine for data engineering, stream processing and data
    science workloads. Each Spark application uses a construct called the
    SparkContext, which is the application’s connection or entry point to the
    Spark engine. Each Spark application will have its own SparkContext.
   
    Livy enables clients to interact with one or more Spark sessions through the
    Livy Server, which acts as a proxy layer. Livy Clients have fine grained
    control over the lifecycle of the Spark sessions, as well as the ability to
    submit jobs and retrieve results, all over HTTP.  Clients have two modes of
    interaction: RPC Client API, available in Java and Python, which allows
    results to be retrieved as Java or Python objects. The serialization and
    deserialization of the results is handled by the Livy framework.  HTTP based
    API that allows submission of code snippets, and retrieval of the results in
    different formats.
   
    Multi-tenant resource allocation and security: Livy enables multiple
    independent Spark sessions to be managed simultaneously. Multiple clients
    can also interact simultaneously with the same Spark session and share the
    resources of that Spark session. Livy can also enforce secure, authenticated
    communication between the clients and their respective Spark sessions.
   
    More information on Livy can be found at the existing open source website:
    http://livy.io/
   
    = Rationale =
   
    Users want to use Spark’s powerful processing engine and API
    as the data processing backend for interactive applications. However, the
    job submission and application interaction mechanisms built into Apache
    Spark are insufficient and cumbersome for multi-user interactive
    applications.
   
    The primary mechanism for applications to submit Spark jobs is via
    spark-submit
    (http://spark.apache.org/docs/latest/submitting-applications.html), which is
    available as a command line tool as well as a programmatic API. However,
    spark-submit has the following limitations that make it difficult to build
    interactive applications: It is slow: each invocation of spark-submit
    involves a setup phase where cluster resources are acquired, new processes
    are forked, etc. This setup phase runs for many seconds, or even minutes,
    and hence is too slow for interactive applications.  It is cumbersome and
    lacks flexibility: application code and dependencies have to be pre-compiled
    and submitted as jars, and can not be submitted interactively.
   
    Apache Spark comes with an ODBC/JDBC server, which can be used to submit SQL
    queries to Spark. However, this solution is limited to SQL and does not
    allow the client to leverage the rest of the Spark API, such as RDDs, MLlib
    and Streaming.
   
    A third way of using Spark is via its command-line shell, which allows the
    interactive submission of snippets of Spark code. However, the shell entails
    running Spark code on the client machine and hence is not a viable mechanism
    for remote clients to submit Spark jobs.
   
    Livy solves the limitations of the above three mechanisms, and provides the
    full Spark API as a multi-tenant service to remote clients.
   
    Since the open source release of Livy in late 2015, we have seen tremendous
    interest among a diverse set of application developers and ISVs that want to
    build applications with Apache Spark. To make Livy a robust and flexible
    solution that will enable a broad and growing set of applications, it is
    important to grow a large and varied community of contributors.
   
    = Initial Goals =
   
    Move existing codebase, website, documentation and mailing
    lists to Apache-hosted infrastructure Work with the infrastructure team to
    implement and approve our code review, build, and testing workflows in the
    context of the ASF Incremental development and releases per Apache
    guidelines
   
    = Current Status =
   
    The Livy project began at Cloudera, as a part of the Hue
    project. Cloudera soon realized the broad applicability of Livy, and
    separated it out into an independent project in Nov 2015.
   
    == Releases ==
   
    Livy has undergone two public releases, tagged here:
   
    * https://github.com/cloudera/livy/releases/tag/v0.2.0
    * https://github.com/cloudera/livy/releases/tag/v0.3.0
   
    Tarballs and zip files were created for each release and hosted on github.
    Upon joining the incubator, we will adopt a more typical ASF release
    process.
   
    == Source ==
   
    Livy’s source is currently hosted on Github at:
   
    https://github.com/cloudera/livy
   
    This repository will be transitioned to Apache’s git hosting during
    incubation.
   
    == Code review ==
   
    Livy’s code reviews are currently public and hosted on
    github as pull request reviews at: https://github.com/cloudera/livy/pulls
    The Livy developer community so far is happy with github pull request
    reviews and hopes to continue this after being admitted to the ASF.
   
    == Issue Tracking ==
   
    Livy’s bug and feature tracking is hosted on JIRA at:
    https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
    contains bugs and development discussion dating back 1 year and will provide
    an initial seed for the ASF JIRA
   
    == Community Discussion ==
   
    Livy has several public discussion forums:
   
    * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
    * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user
   
    == Development Practices ==
   
    The Livy project follows a review before commit philosophy. Every commit
    automatically runs through the unit tests and generates coverage reports
    presented as a pull request comment. Our experience with this process leads
    us to believe that it helps ease new contributors into the project. They get
    feedback quickly on common mistakes, lowering the burden on reviewers. Those
    same reviewers get to lead by example, showing the new contributors that we
    value feedback within our community even when changes are done by more
    experienced folks.
   
    == Meritocracy ==
   
    We believe strongly in meritocracy when electing committers and PMC members.
    In the past few months, the project has added two new committers from two
    different organisations, in recognition of their significant contributions
    to the project. We will encourage contributions and participation of all
    types, and ensure that contributors are appropriately recognized.
   
    == Community ==
   
    Though Livy is relatively new as a standalone open source project, it has
    already seen promising growth in its community across several organizations:
    Cloudera is the original development sponsor for Livy Microsoft pushed the
    development of the interpreter fixing high availability issues and adding
    additional features.  Hortonworks has contributed the security features to
    Livy allowing kerberos and impersonation to work with Spark IBM is starting
    to make contributions to the Livy project A number of other patches
    contributed by community members
   
    Livy currently relies on Google Groups for mailing lists. These lists have
    been active since the end of 2015/start of 2016. Currently, Livy’s user
    mailing list has 173 subscribers and has hosted a total of 227 topic
    threads. Livy’s developer list has 49 subscribers and has hosted 79 topic
    threads.
   
    == Core Developers ==
   
    The early contributions to Livy were made by Cloudera engineers. In 2016,
    engineers from Microsoft and Hortonworks joined the core developer
    community.
   
    == Alignment ==
   
    Livy is built upon Apache Spark, and other Apache projects like Apache
    Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
    community connections combined with our focus on development practices that
    emphasize community engagement with a path to meritocratic recognition
    naturally align us with the ASF.
   
    = Known Risks =
    == Orphaned Products ==
   
    The risk of Livy being abandoned is low because it is supported by three
    major big-data software vendors.  Moreover, Livy is already used to power
    multiple releases of services and products used in production.
   
    == Inexperience with Open Source ==
   
    Several of the initial committers are experienced open source developers,
    several being committers and/or PMC members on other ASF projects (Spark,
    YARN).
   
    == Homogenous Developers ==
   
    The project already has a diverse developer base. It has contributions from
    3 major organisations (Cloudera, Microsoft and Hortonworks), and is used in
    diverse applications, in diverse settings (On-Prem and Cloud).
   
    == Reliance on salaried Developers ==
   
    The existing contributors to the Livy project have been made by salaried
    engineers from Cloudera, Microsoft and Hortonworks. Since there are three
    major organisations involved, the risk of reliance on a single group of
    salaried developers is mitigated. The Livy user base is diverse, with users
    from across the globe, including users from academic settings. We aim to
    further diversify the Livy user and contributor base.
   
    == Relationships with other Apache projects ==
   
    Livy is closely tied to the Apache Spark project and currently addresses the
    scenarios for a REST based batch and interactive gateway for Spark jobs on
    YARN. Given the growing number of integrations with Livy, keeping it outside
    of Apache Spark aligns with the desire of the Apache Spark community to
    reduce the number of external dependencies in the Spark project.
    Specifically, the Apache Spark community has previously expressed a desire
    to keep job servers independent from the project.<<FootNote(See, for
    example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
    Furthermore, while Livy common usage is closely tied to Spark deployments
    right now, its core building blocks can be reused elsewhere.  Livy’s Remote
    REPL could be used as a library for interactive scenarios in non-Spark
    projects. In the future, integrations with cluster managers like Apache
    Mesos and others could also be added.
   
    The features provided by Livy have already been integrated with existing
    projects like Jupyter and Apache Zeppelin for their interactive Spark use
    cases. This validates the need for a project like Livy and provides an
    active downstream user base that the Livy community can interact with to
    seed future interest in the project.
   
    Livy serves a similar purpose to Apache Toree (incubating) but differs in
    making session management, security and impersonation a focal design point.
   
    == An Excessive Fascination with the Apache Brand ==
   
    The primary motivation for submitting Livy to the ASF is to grow a diverse
    and strong community. We wish to encourage diverse organisations, including
    ISVs, to adopt Livy and contribute to Livy without any concerns about
    ownership or licensing.
   
    = Documentation =
   
    Documentation can be found on the Livy website http://livy.io/ The Livy web
    site is version controlled on the ‘gh-pages’ branch of the above repository
    Additional documentation is provided on the github wiki:
    https://github.com/cloudera/livy/wiki APis are documented within the source
    code as JavaDoc style documentation comments.
   
    = Initial Source =
   
    The initial source code for Livy is hosted at
   
    https://github.com/cloudera/livy 
   
    = Source and Intellectual Property submission plan =
   
    The Livy codebase and web site is currently hosted on GitHub and will be
    transitioned to the ASF repositories during incubation. Livy is already
    licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
    CCLAs from all committers.  There are, however, some contributions recently
    from authors that have not signed the CCLA and ICLA. If necessary for a
    successful SGA, we’ll seek the necessary documentation or replace the
    contributions.
   
    The “Livy” name is not a registered trademark. We will need to do a
    trademark search and make sure it is available for the Apache Foundation
    prior to graduation.
   
    Cloudera currently owns the domain name: http://livy.io/ which will be
    transferred to the ASF and redirected to the official page during
    incubation.
   
    = External Dependencies =
   
    The list below covers the non-Apache dependencies of the project and their
    licenses.
   
     * Jetty: Apache 2.0
     * Dropwizard Metrics: Apache 2.0
     * FasterXML Jackson: Apache 2.0
     * Netty: Apache 2.0
     * Scala: BSD
     * Py4J: BSD
     * Scalatra: BSD
   
    Build/test-only dependencies:
   
     * Mockito: MIT
     * JUnit: Eclipse
   
    = Required Resources =
    == Mailing Lists ==
   
     * [hidden email] (PPMC)
     * [hidden email] (dev mailing list)
     * [hidden email] (User questions)
     * [hidden email] (subscribers shouldn’t be able to post)
     * [hidden email] (subscribers shouldn’t be able to post)
   
    == Git Repository ==
   
    git://git.apache.org/livy
   
    == Issue Tracking ==
   
    We would like to import our current JIRA project into the ASF JIRA, such
    that our historical commit message and code comments continue to reference
    the appropriate bug numbers.
   
    = Initial Committers =
   
     * Marcelo Vanzin ([hidden email])
     * Alex Man ([hidden email])
     * Jeff Zhang ([hidden email])
     * Saisai Shao ([hidden email])
     * Kostas Sakellis ([hidden email])
   
    = Affiliations =
   
    The initial set of committers includes people employed by Cloudera and
    Hortonworks as well as one person currently unaffiliated with an employer.
   
    = Additional Interested Contributors =
   
    Those interested in getting involved with the project as we enter incubation
    are encourage to list themselves here.
   
     * < add here >
   
    = Sponsors =
    == Champion ==
   
     * Sean Busbey ([hidden email])
   
    == Nominated Mentors ==
   
     * Bikas Saha ([hidden email])
     * Brock Noland ([hidden email])
   
    == Sponsoring Entity ==
   
    We ask that the Incubator PMC sponsor this proposal.
   
   
    ---------------------------------------------------------------------
    To unsubscribe, e-mail: [hidden email]
    For additional commands, e-mail: [hidden email]
   
   


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Brock Noland
Great to see!

+1

On Fri, May 19, 2017 at 7:24 PM, William GUO <[hidden email]> wrote:

> +1
>
> Griffin needs Livy to access Spark context.
>
>
> Thanks,
> William
>
> On 5/20/17, 7:45 AM, "Sean Busbey" <[hidden email]> wrote:
>
>     Dear Apache Incubator Community,
>
>     I'm excited to present for discussion a proposal to move Livy into
>     incubation. Livy is web service that exposes a REST interface for
> managing
>     long running Apache Spark contexts in your cluster. With Livy, new
>     applications can be built on top of Apache Spark that require fine
> grained
>     interaction with many Spark contexts.
>
>     The proposal is on the wiki at the following page as well as copied in
> the
>     email below:
>
>     https://wiki.apache.org/incubator/LivyProposal
>
>     In addition to welcoming feedback on the proposal, we are actively
> seeking
>     one or more additional mentors. We also have included a section for
>     interested folks to ensure they get added to the mailing lists,
> presuming
>     Livy gets accepted for incubation.
>
>     ---- LivyProposal
>
>     = Abstract =
>
>     Livy is web service that exposes a REST interface for managing
>     long running Apache Spark contexts in your cluster. With Livy, new
>     applications can be built on top of Apache Spark that require fine
> grained
>     interaction with many Spark contexts.
>
>     = Proposal =
>
>     Livy is an open-source REST service for Apache Spark. Livy
>     enables applications to submit Spark applications and retrieve results
>     without a co-location requirement on the Spark cluster.
>
>     We propose to contribute the Livy codebase and associated artifacts
> (e.g.
>     documentation, web-site context etc) to the Apache Software Foundation.
>
>     = Background =
>
>     Apache Spark is a fast and general purpose distributed
>     compute engine, with a versatile API. It enables processing of large
>     quantities of static data distributed over a cluster of machines, as
> well as
>     processing of continuous streams of data. It is the preferred
> distributed
>     data processing engine for data engineering, stream processing and data
>     science workloads. Each Spark application uses a construct called the
>     SparkContext, which is the application’s connection or entry point
> to the
>     Spark engine. Each Spark application will have its own SparkContext.
>
>     Livy enables clients to interact with one or more Spark sessions
> through the
>     Livy Server, which acts as a proxy layer. Livy Clients have fine
> grained
>     control over the lifecycle of the Spark sessions, as well as the
> ability to
>     submit jobs and retrieve results, all over HTTP.  Clients have two
> modes of
>     interaction: RPC Client API, available in Java and Python, which allows
>     results to be retrieved as Java or Python objects. The serialization
> and
>     deserialization of the results is handled by the Livy framework.  HTTP
> based
>     API that allows submission of code snippets, and retrieval of the
> results in
>     different formats.
>
>     Multi-tenant resource allocation and security: Livy enables multiple
>     independent Spark sessions to be managed simultaneously. Multiple
> clients
>     can also interact simultaneously with the same Spark session and share
> the
>     resources of that Spark session. Livy can also enforce secure,
> authenticated
>     communication between the clients and their respective Spark sessions.
>
>     More information on Livy can be found at the existing open source
> website:
>     http://livy.io/
>
>     = Rationale =
>
>     Users want to use Spark’s powerful processing engine and API
>     as the data processing backend for interactive applications. However,
> the
>     job submission and application interaction mechanisms built into Apache
>     Spark are insufficient and cumbersome for multi-user interactive
>     applications.
>
>     The primary mechanism for applications to submit Spark jobs is via
>     spark-submit
>     (http://spark.apache.org/docs/latest/submitting-applications.html),
> which is
>     available as a command line tool as well as a programmatic API.
> However,
>     spark-submit has the following limitations that make it difficult to
> build
>     interactive applications: It is slow: each invocation of spark-submit
>     involves a setup phase where cluster resources are acquired, new
> processes
>     are forked, etc. This setup phase runs for many seconds, or even
> minutes,
>     and hence is too slow for interactive applications.  It is cumbersome
> and
>     lacks flexibility: application code and dependencies have to be
> pre-compiled
>     and submitted as jars, and can not be submitted interactively.
>
>     Apache Spark comes with an ODBC/JDBC server, which can be used to
> submit SQL
>     queries to Spark. However, this solution is limited to SQL and does not
>     allow the client to leverage the rest of the Spark API, such as RDDs,
> MLlib
>     and Streaming.
>
>     A third way of using Spark is via its command-line shell, which allows
> the
>     interactive submission of snippets of Spark code. However, the shell
> entails
>     running Spark code on the client machine and hence is not a viable
> mechanism
>     for remote clients to submit Spark jobs.
>
>     Livy solves the limitations of the above three mechanisms, and
> provides the
>     full Spark API as a multi-tenant service to remote clients.
>
>     Since the open source release of Livy in late 2015, we have seen
> tremendous
>     interest among a diverse set of application developers and ISVs that
> want to
>     build applications with Apache Spark. To make Livy a robust and
> flexible
>     solution that will enable a broad and growing set of applications, it
> is
>     important to grow a large and varied community of contributors.
>
>     = Initial Goals =
>
>     Move existing codebase, website, documentation and mailing
>     lists to Apache-hosted infrastructure Work with the infrastructure
> team to
>     implement and approve our code review, build, and testing workflows in
> the
>     context of the ASF Incremental development and releases per Apache
>     guidelines
>
>     = Current Status =
>
>     The Livy project began at Cloudera, as a part of the Hue
>     project. Cloudera soon realized the broad applicability of Livy, and
>     separated it out into an independent project in Nov 2015.
>
>     == Releases ==
>
>     Livy has undergone two public releases, tagged here:
>
>     * https://github.com/cloudera/livy/releases/tag/v0.2.0
>     * https://github.com/cloudera/livy/releases/tag/v0.3.0
>
>     Tarballs and zip files were created for each release and hosted on
> github.
>     Upon joining the incubator, we will adopt a more typical ASF release
>     process.
>
>     == Source ==
>
>     Livy’s source is currently hosted on Github at:
>
>     https://github.com/cloudera/livy
>
>     This repository will be transitioned to Apache’s git hosting during
>     incubation.
>
>     == Code review ==
>
>     Livy’s code reviews are currently public and hosted on
>     github as pull request reviews at: https://github.com/cloudera/
> livy/pulls
>     The Livy developer community so far is happy with github pull request
>     reviews and hopes to continue this after being admitted to the ASF.
>
>     == Issue Tracking ==
>
>     Livy’s bug and feature tracking is hosted on JIRA at:
>     https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
>     contains bugs and development discussion dating back 1 year and will
> provide
>     an initial seed for the ASF JIRA
>
>     == Community Discussion ==
>
>     Livy has several public discussion forums:
>
>     * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
>     * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user
>
>     == Development Practices ==
>
>     The Livy project follows a review before commit philosophy. Every
> commit
>     automatically runs through the unit tests and generates coverage
> reports
>     presented as a pull request comment. Our experience with this process
> leads
>     us to believe that it helps ease new contributors into the project.
> They get
>     feedback quickly on common mistakes, lowering the burden on reviewers.
> Those
>     same reviewers get to lead by example, showing the new contributors
> that we
>     value feedback within our community even when changes are done by more
>     experienced folks.
>
>     == Meritocracy ==
>
>     We believe strongly in meritocracy when electing committers and PMC
> members.
>     In the past few months, the project has added two new committers from
> two
>     different organisations, in recognition of their significant
> contributions
>     to the project. We will encourage contributions and participation of
> all
>     types, and ensure that contributors are appropriately recognized.
>
>     == Community ==
>
>     Though Livy is relatively new as a standalone open source project, it
> has
>     already seen promising growth in its community across several
> organizations:
>     Cloudera is the original development sponsor for Livy Microsoft pushed
> the
>     development of the interpreter fixing high availability issues and
> adding
>     additional features.  Hortonworks has contributed the security
> features to
>     Livy allowing kerberos and impersonation to work with Spark IBM is
> starting
>     to make contributions to the Livy project A number of other patches
>     contributed by community members
>
>     Livy currently relies on Google Groups for mailing lists. These lists
> have
>     been active since the end of 2015/start of 2016. Currently, Livy’s
> user
>     mailing list has 173 subscribers and has hosted a total of 227 topic
>     threads. Livy’s developer list has 49 subscribers and has hosted 79
> topic
>     threads.
>
>     == Core Developers ==
>
>     The early contributions to Livy were made by Cloudera engineers. In
> 2016,
>     engineers from Microsoft and Hortonworks joined the core developer
>     community.
>
>     == Alignment ==
>
>     Livy is built upon Apache Spark, and other Apache projects like Apache
>     Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
>     community connections combined with our focus on development practices
> that
>     emphasize community engagement with a path to meritocratic recognition
>     naturally align us with the ASF.
>
>     = Known Risks =
>     == Orphaned Products ==
>
>     The risk of Livy being abandoned is low because it is supported by
> three
>     major big-data software vendors.  Moreover, Livy is already used to
> power
>     multiple releases of services and products used in production.
>
>     == Inexperience with Open Source ==
>
>     Several of the initial committers are experienced open source
> developers,
>     several being committers and/or PMC members on other ASF projects
> (Spark,
>     YARN).
>
>     == Homogenous Developers ==
>
>     The project already has a diverse developer base. It has contributions
> from
>     3 major organisations (Cloudera, Microsoft and Hortonworks), and is
> used in
>     diverse applications, in diverse settings (On-Prem and Cloud).
>
>     == Reliance on salaried Developers ==
>
>     The existing contributors to the Livy project have been made by
> salaried
>     engineers from Cloudera, Microsoft and Hortonworks. Since there are
> three
>     major organisations involved, the risk of reliance on a single group of
>     salaried developers is mitigated. The Livy user base is diverse, with
> users
>     from across the globe, including users from academic settings. We aim
> to
>     further diversify the Livy user and contributor base.
>
>     == Relationships with other Apache projects ==
>
>     Livy is closely tied to the Apache Spark project and currently
> addresses the
>     scenarios for a REST based batch and interactive gateway for Spark
> jobs on
>     YARN. Given the growing number of integrations with Livy, keeping it
> outside
>     of Apache Spark aligns with the desire of the Apache Spark community to
>     reduce the number of external dependencies in the Spark project.
>     Specifically, the Apache Spark community has previously expressed a
> desire
>     to keep job servers independent from the project.<<FootNote(See, for
>     example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
>     Furthermore, while Livy common usage is closely tied to Spark
> deployments
>     right now, its core building blocks can be reused elsewhere.  Livy’s
> Remote
>     REPL could be used as a library for interactive scenarios in non-Spark
>     projects. In the future, integrations with cluster managers like Apache
>     Mesos and others could also be added.
>
>     The features provided by Livy have already been integrated with
> existing
>     projects like Jupyter and Apache Zeppelin for their interactive Spark
> use
>     cases. This validates the need for a project like Livy and provides an
>     active downstream user base that the Livy community can interact with
> to
>     seed future interest in the project.
>
>     Livy serves a similar purpose to Apache Toree (incubating) but differs
> in
>     making session management, security and impersonation a focal design
> point.
>
>     == An Excessive Fascination with the Apache Brand ==
>
>     The primary motivation for submitting Livy to the ASF is to grow a
> diverse
>     and strong community. We wish to encourage diverse organisations,
> including
>     ISVs, to adopt Livy and contribute to Livy without any concerns about
>     ownership or licensing.
>
>     = Documentation =
>
>     Documentation can be found on the Livy website http://livy.io/ The
> Livy web
>     site is version controlled on the ‘gh-pages’ branch of the above
> repository
>     Additional documentation is provided on the github wiki:
>     https://github.com/cloudera/livy/wiki APis are documented within the
> source
>     code as JavaDoc style documentation comments.
>
>     = Initial Source =
>
>     The initial source code for Livy is hosted at
>
>     https://github.com/cloudera/livy
>
>     = Source and Intellectual Property submission plan =
>
>     The Livy codebase and web site is currently hosted on GitHub and will
> be
>     transitioned to the ASF repositories during incubation. Livy is already
>     licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
>     CCLAs from all committers.  There are, however, some contributions
> recently
>     from authors that have not signed the CCLA and ICLA. If necessary for a
>     successful SGA, we’ll seek the necessary documentation or replace the
>     contributions.
>
>     The “Livy† name is not a registered trademark. We will need to do a
>     trademark search and make sure it is available for the Apache
> Foundation
>     prior to graduation.
>
>     Cloudera currently owns the domain name: http://livy.io/ which will be
>     transferred to the ASF and redirected to the official page during
>     incubation.
>
>     = External Dependencies =
>
>     The list below covers the non-Apache dependencies of the project and
> their
>     licenses.
>
>      * Jetty: Apache 2.0
>      * Dropwizard Metrics: Apache 2.0
>      * FasterXML Jackson: Apache 2.0
>      * Netty: Apache 2.0
>      * Scala: BSD
>      * Py4J: BSD
>      * Scalatra: BSD
>
>     Build/test-only dependencies:
>
>      * Mockito: MIT
>      * JUnit: Eclipse
>
>     = Required Resources =
>     == Mailing Lists ==
>
>      * [hidden email] (PPMC)
>      * [hidden email] (dev mailing list)
>      * [hidden email] (User questions)
>      * [hidden email] (subscribers shouldn’t be able
> to post)
>      * [hidden email] (subscribers shouldn’t be able
> to post)
>
>     == Git Repository ==
>
>     git://git.apache.org/livy
>
>     == Issue Tracking ==
>
>     We would like to import our current JIRA project into the ASF JIRA,
> such
>     that our historical commit message and code comments continue to
> reference
>     the appropriate bug numbers.
>
>     = Initial Committers =
>
>      * Marcelo Vanzin ([hidden email])
>      * Alex Man ([hidden email])
>      * Jeff Zhang ([hidden email])
>      * Saisai Shao ([hidden email])
>      * Kostas Sakellis ([hidden email])
>
>     = Affiliations =
>
>     The initial set of committers includes people employed by Cloudera and
>     Hortonworks as well as one person currently unaffiliated with an
> employer.
>
>     = Additional Interested Contributors =
>
>     Those interested in getting involved with the project as we enter
> incubation
>     are encourage to list themselves here.
>
>      * < add here >
>
>     = Sponsors =
>     == Champion ==
>
>      * Sean Busbey ([hidden email])
>
>     == Nominated Mentors ==
>
>      * Bikas Saha ([hidden email])
>      * Brock Noland ([hidden email])
>
>     == Sponsoring Entity ==
>
>     We ask that the Incubator PMC sponsor this proposal.
>
>
>     ---------------------------------------------------------------------
>     To unsubscribe, e-mail: [hidden email]
>     For additional commands, e-mail: [hidden email]
>
>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

John D. Ament-2
In reply to this post by Sean Busbey-2
Sean,

On Fri, May 19, 2017 at 7:45 PM Sean Busbey <[hidden email]> wrote:

> Dear Apache Incubator Community,
>
> I'm excited to present for discussion a proposal to move Livy into
> incubation. Livy is web service that exposes a REST interface for managing
> long running Apache Spark contexts in your cluster. With Livy, new
> applications can be built on top of Apache Spark that require fine grained
> interaction with many Spark contexts.
>
> The proposal is on the wiki at the following page as well as copied in the
> email below:
>
> https://wiki.apache.org/incubator/LivyProposal
>
> In addition to welcoming feedback on the proposal, we are actively seeking
> one or more additional mentors. We also have included a section for
> interested folks to ensure they get added to the mailing lists, presuming
> Livy gets accepted for incubation.
>
> ---- LivyProposal
>
> = Abstract =
>
> Livy is web service that exposes a REST interface for managing
> long running Apache Spark contexts in your cluster. With Livy, new
> applications can be built on top of Apache Spark that require fine grained
> interaction with many Spark contexts.
>
> = Proposal =
>
> Livy is an open-source REST service for Apache Spark. Livy
> enables applications to submit Spark applications and retrieve results
> without a co-location requirement on the Spark cluster.
>
> We propose to contribute the Livy codebase and associated artifacts (e.g.
> documentation, web-site context etc) to the Apache Software Foundation.
>
> = Background =
>
> Apache Spark is a fast and general purpose distributed
> compute engine, with a versatile API. It enables processing of large
> quantities of static data distributed over a cluster of machines, as well
> as
> processing of continuous streams of data. It is the preferred distributed
> data processing engine for data engineering, stream processing and data
> science workloads. Each Spark application uses a construct called the
> SparkContext, which is the application’s connection or entry point to the
> Spark engine. Each Spark application will have its own SparkContext.
>
> Livy enables clients to interact with one or more Spark sessions through
> the
> Livy Server, which acts as a proxy layer. Livy Clients have fine grained
> control over the lifecycle of the Spark sessions, as well as the ability to
> submit jobs and retrieve results, all over HTTP.  Clients have two modes of
> interaction: RPC Client API, available in Java and Python, which allows
> results to be retrieved as Java or Python objects. The serialization and
> deserialization of the results is handled by the Livy framework.  HTTP
> based
> API that allows submission of code snippets, and retrieval of the results
> in
> different formats.
>
> Multi-tenant resource allocation and security: Livy enables multiple
> independent Spark sessions to be managed simultaneously. Multiple clients
> can also interact simultaneously with the same Spark session and share the
> resources of that Spark session. Livy can also enforce secure,
> authenticated
> communication between the clients and their respective Spark sessions.
>
> More information on Livy can be found at the existing open source website:
> http://livy.io/
>
> = Rationale =
>
> Users want to use Spark’s powerful processing engine and API
> as the data processing backend for interactive applications. However, the
> job submission and application interaction mechanisms built into Apache
> Spark are insufficient and cumbersome for multi-user interactive
> applications.
>
> The primary mechanism for applications to submit Spark jobs is via
> spark-submit
> (http://spark.apache.org/docs/latest/submitting-applications.html), which
> is
> available as a command line tool as well as a programmatic API. However,
> spark-submit has the following limitations that make it difficult to build
> interactive applications: It is slow: each invocation of spark-submit
> involves a setup phase where cluster resources are acquired, new processes
> are forked, etc. This setup phase runs for many seconds, or even minutes,
> and hence is too slow for interactive applications.  It is cumbersome and
> lacks flexibility: application code and dependencies have to be
> pre-compiled
> and submitted as jars, and can not be submitted interactively.
>
> Apache Spark comes with an ODBC/JDBC server, which can be used to submit
> SQL
> queries to Spark. However, this solution is limited to SQL and does not
> allow the client to leverage the rest of the Spark API, such as RDDs, MLlib
> and Streaming.
>
> A third way of using Spark is via its command-line shell, which allows the
> interactive submission of snippets of Spark code. However, the shell
> entails
> running Spark code on the client machine and hence is not a viable
> mechanism
> for remote clients to submit Spark jobs.
>
> Livy solves the limitations of the above three mechanisms, and provides the
> full Spark API as a multi-tenant service to remote clients.
>
> Since the open source release of Livy in late 2015, we have seen tremendous
> interest among a diverse set of application developers and ISVs that want
> to
> build applications with Apache Spark. To make Livy a robust and flexible
> solution that will enable a broad and growing set of applications, it is
> important to grow a large and varied community of contributors.
>
> = Initial Goals =
>
> Move existing codebase, website, documentation and mailing
> lists to Apache-hosted infrastructure Work with the infrastructure team to
> implement and approve our code review, build, and testing workflows in the
> context of the ASF Incremental development and releases per Apache
> guidelines
>
> = Current Status =
>
> The Livy project began at Cloudera, as a part of the Hue
> project. Cloudera soon realized the broad applicability of Livy, and
> separated it out into an independent project in Nov 2015.
>
> == Releases ==
>
> Livy has undergone two public releases, tagged here:
>
> * https://github.com/cloudera/livy/releases/tag/v0.2.0
> * https://github.com/cloudera/livy/releases/tag/v0.3.0
>
> Tarballs and zip files were created for each release and hosted on github.
> Upon joining the incubator, we will adopt a more typical ASF release
> process.
>
> == Source ==
>
> Livy’s source is currently hosted on Github at:
>
> https://github.com/cloudera/livy
>
> This repository will be transitioned to Apache’s git hosting during
> incubation.
>
> == Code review ==
>
> Livy’s code reviews are currently public and hosted on
> github as pull request reviews at: https://github.com/cloudera/livy/pulls
> The Livy developer community so far is happy with github pull request
> reviews and hopes to continue this after being admitted to the ASF.
>
> == Issue Tracking ==
>
> Livy’s bug and feature tracking is hosted on JIRA at:
> https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
> contains bugs and development discussion dating back 1 year and will
> provide
> an initial seed for the ASF JIRA
>
> == Community Discussion ==
>
> Livy has several public discussion forums:
>
> * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
> * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user
>
> == Development Practices ==
>
> The Livy project follows a review before commit philosophy. Every commit
> automatically runs through the unit tests and generates coverage reports
> presented as a pull request comment. Our experience with this process leads
> us to believe that it helps ease new contributors into the project. They
> get
> feedback quickly on common mistakes, lowering the burden on reviewers.
> Those
> same reviewers get to lead by example, showing the new contributors that we
> value feedback within our community even when changes are done by more
> experienced folks.
>
> == Meritocracy ==
>
> We believe strongly in meritocracy when electing committers and PMC
> members.
> In the past few months, the project has added two new committers from two
> different organisations, in recognition of their significant contributions
> to the project. We will encourage contributions and participation of all
> types, and ensure that contributors are appropriately recognized.
>
> == Community ==
>
> Though Livy is relatively new as a standalone open source project, it has
> already seen promising growth in its community across several
> organizations:
> Cloudera is the original development sponsor for Livy Microsoft pushed the
> development of the interpreter fixing high availability issues and adding
> additional features.  Hortonworks has contributed the security features to
> Livy allowing kerberos and impersonation to work with Spark IBM is starting
> to make contributions to the Livy project A number of other patches
> contributed by community members
>
> Livy currently relies on Google Groups for mailing lists. These lists have
> been active since the end of 2015/start of 2016. Currently, Livy’s user
> mailing list has 173 subscribers and has hosted a total of 227 topic
> threads. Livy’s developer list has 49 subscribers and has hosted 79 topic
> threads.
>
> == Core Developers ==
>
> The early contributions to Livy were made by Cloudera engineers. In 2016,
> engineers from Microsoft and Hortonworks joined the core developer
> community.
>
> == Alignment ==
>
> Livy is built upon Apache Spark, and other Apache projects like Apache
> Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
> community connections combined with our focus on development practices that
> emphasize community engagement with a path to meritocratic recognition
> naturally align us with the ASF.
>
> = Known Risks =
> == Orphaned Products ==
>
> The risk of Livy being abandoned is low because it is supported by three
> major big-data software vendors.  Moreover, Livy is already used to power
> multiple releases of services and products used in production.
>
> == Inexperience with Open Source ==
>
> Several of the initial committers are experienced open source developers,
> several being committers and/or PMC members on other ASF projects (Spark,
> YARN).
>
> == Homogenous Developers ==
>
> The project already has a diverse developer base. It has contributions from
> 3 major organisations (Cloudera, Microsoft and Hortonworks), and is used in
> diverse applications, in diverse settings (On-Prem and Cloud).
>
> == Reliance on salaried Developers ==
>
> The existing contributors to the Livy project have been made by salaried
> engineers from Cloudera, Microsoft and Hortonworks. Since there are three
> major organisations involved, the risk of reliance on a single group of
> salaried developers is mitigated. The Livy user base is diverse, with users
> from across the globe, including users from academic settings. We aim to
> further diversify the Livy user and contributor base.
>
>
There's a disconnect between this paragraph and the initial committers
list.  Specifically, no one from Microsoft is represented (as best as I can
tell).


> == Relationships with other Apache projects ==
>
> Livy is closely tied to the Apache Spark project and currently addresses
> the
> scenarios for a REST based batch and interactive gateway for Spark jobs on
> YARN. Given the growing number of integrations with Livy, keeping it
> outside
> of Apache Spark aligns with the desire of the Apache Spark community to
> reduce the number of external dependencies in the Spark project.
> Specifically, the Apache Spark community has previously expressed a desire
> to keep job servers independent from the project.<<FootNote(See, for
> example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
> Furthermore, while Livy common usage is closely tied to Spark deployments
> right now, its core building blocks can be reused elsewhere.  Livy’s Remote
> REPL could be used as a library for interactive scenarios in non-Spark
> projects. In the future, integrations with cluster managers like Apache
> Mesos and others could also be added.
>
> The features provided by Livy have already been integrated with existing
> projects like Jupyter and Apache Zeppelin for their interactive Spark use
> cases. This validates the need for a project like Livy and provides an
> active downstream user base that the Livy community can interact with to
> seed future interest in the project.
>
> Livy serves a similar purpose to Apache Toree (incubating) but differs in
> making session management, security and impersonation a focal design point.
>
> == An Excessive Fascination with the Apache Brand ==
>
> The primary motivation for submitting Livy to the ASF is to grow a diverse
> and strong community. We wish to encourage diverse organisations, including
> ISVs, to adopt Livy and contribute to Livy without any concerns about
> ownership or licensing.
>
> = Documentation =
>
> Documentation can be found on the Livy website http://livy.io/ The Livy
> web
> site is version controlled on the ‘gh-pages’ branch of the above repository
> Additional documentation is provided on the github wiki:
> https://github.com/cloudera/livy/wiki APis are documented within the
> source
> code as JavaDoc style documentation comments.
>
> = Initial Source =
>
> The initial source code for Livy is hosted at
>
> https://github.com/cloudera/livy
>
> = Source and Intellectual Property submission plan =
>
> The Livy codebase and web site is currently hosted on GitHub and will be
> transitioned to the ASF repositories during incubation. Livy is already
> licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
> CCLAs from all committers.  There are, however, some contributions recently
> from authors that have not signed the CCLA and ICLA. If necessary for a
> successful SGA, we’ll seek the necessary documentation or replace the
> contributions.
>
> The “Livy” name is not a registered trademark. We will need to do a
> trademark search and make sure it is available for the Apache Foundation
> prior to graduation.
>
>

> Cloudera currently owns the domain name: http://livy.io/ which will be
> transferred to the ASF and redirected to the official page during
> incubation.
>
>
I'm assuming that the incoming project is aware that we expect the main dev
landing page to be livy.incubator.apache.org .  We will want to track this
as a project specific item.


> = External Dependencies =
>
> The list below covers the non-Apache dependencies of the project and their
> licenses.
>
>  * Jetty: Apache 2.0
>  * Dropwizard Metrics: Apache 2.0
>  * FasterXML Jackson: Apache 2.0
>  * Netty: Apache 2.0
>  * Scala: BSD
>  * Py4J: BSD
>  * Scalatra: BSD
>
> Build/test-only dependencies:
>
>  * Mockito: MIT
>  * JUnit: Eclipse
>
> = Required Resources =
> == Mailing Lists ==
>
>  * [hidden email] (PPMC)
>  * [hidden email] (dev mailing list)
>  * [hidden email] (User questions)
>  * [hidden email] (subscribers shouldn’t be able to
> post)
>  * [hidden email] (subscribers shouldn’t be able to
> post)
>
> == Git Repository ==
>
> git://git.apache.org/livy
>
>
Just to confirm - it'll be incubator-livy, not just livy.


> == Issue Tracking ==
>
> We would like to import our current JIRA project into the ASF JIRA, such
> that our historical commit message and code comments continue to reference
> the appropriate bug numbers.
>
>
I would recommend reaching out to infra to see if the import is possible
before voting on the project.  Otherwise you'll need to list out an la


> = Initial Committers =
>
>  * Marcelo Vanzin ([hidden email])
>  * Alex Man ([hidden email])
>  * Jeff Zhang ([hidden email])
>  * Saisai Shao ([hidden email])
>  * Kostas Sakellis ([hidden email])
>
> = Affiliations =
>
> The initial set of committers includes people employed by Cloudera and
> Hortonworks as well as one person currently unaffiliated with an employer.
>
> = Additional Interested Contributors =
>
> Those interested in getting involved with the project as we enter
> incubation
> are encourage to list themselves here.
>
>  * < add here >
>
> = Sponsors =
> == Champion ==
>
>  * Sean Busbey ([hidden email])
>
> == Nominated Mentors ==
>
>  * Bikas Saha ([hidden email])
>  * Brock Noland ([hidden email])
>

A couple of points:

- Sean, while the champion and mentor roles are separate, we do hope that
all champions will continue on as a mentor.  If this is your intention
please add yourself.
- All mentors must be on the IPMC.  Foundation membership isn't a
requirement, however most people use membership to get access to the IPMC.
If Bikas wants to be a mentor, he'll need to join the IPMC otherwise you'll
need to find 2 mentors.
- Do the proposed mentors have a relationship to the incoming project, e.g.
do they care if it succeeds from a corporate interest standpoint?


>
> == Sponsoring Entity ==
>
> We ask that the Incubator PMC sponsor this proposal.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Luciano Resende
In reply to this post by Sean Busbey-2
+1

Also, I see the proposal is short on mentors, so feel free to include me as
a mentor for the project.

Thanks

On Fri, May 19, 2017 at 4:45 PM Sean Busbey <[hidden email]> wrote:

> Dear Apache Incubator Community,
>
> I'm excited to present for discussion a proposal to move Livy into
> incubation. Livy is web service that exposes a REST interface for managing
> long running Apache Spark contexts in your cluster. With Livy, new
> applications can be built on top of Apache Spark that require fine grained
> interaction with many Spark contexts.
>
> The proposal is on the wiki at the following page as well as copied in the
> email below:
>
> https://wiki.apache.org/incubator/LivyProposal
>
> In addition to welcoming feedback on the proposal, we are actively seeking
> one or more additional mentors. We also have included a section for
> interested folks to ensure they get added to the mailing lists, presuming
> Livy gets accepted for incubation.
>
> ---- LivyProposal
>
> = Abstract =
>
> Livy is web service that exposes a REST interface for managing
> long running Apache Spark contexts in your cluster. With Livy, new
> applications can be built on top of Apache Spark that require fine grained
> interaction with many Spark contexts.
>
> = Proposal =
>
> Livy is an open-source REST service for Apache Spark. Livy
> enables applications to submit Spark applications and retrieve results
> without a co-location requirement on the Spark cluster.
>
> We propose to contribute the Livy codebase and associated artifacts (e.g.
> documentation, web-site context etc) to the Apache Software Foundation.
>
> = Background =
>
> Apache Spark is a fast and general purpose distributed
> compute engine, with a versatile API. It enables processing of large
> quantities of static data distributed over a cluster of machines, as well
> as
> processing of continuous streams of data. It is the preferred distributed
> data processing engine for data engineering, stream processing and data
> science workloads. Each Spark application uses a construct called the
> SparkContext, which is the application’s connection or entry point to the
> Spark engine. Each Spark application will have its own SparkContext.
>
> Livy enables clients to interact with one or more Spark sessions through
> the
> Livy Server, which acts as a proxy layer. Livy Clients have fine grained
> control over the lifecycle of the Spark sessions, as well as the ability to
> submit jobs and retrieve results, all over HTTP.  Clients have two modes of
> interaction: RPC Client API, available in Java and Python, which allows
> results to be retrieved as Java or Python objects. The serialization and
> deserialization of the results is handled by the Livy framework.  HTTP
> based
> API that allows submission of code snippets, and retrieval of the results
> in
> different formats.
>
> Multi-tenant resource allocation and security: Livy enables multiple
> independent Spark sessions to be managed simultaneously. Multiple clients
> can also interact simultaneously with the same Spark session and share the
> resources of that Spark session. Livy can also enforce secure,
> authenticated
> communication between the clients and their respective Spark sessions.
>
> More information on Livy can be found at the existing open source website:
> http://livy.io/
>
> = Rationale =
>
> Users want to use Spark’s powerful processing engine and API
> as the data processing backend for interactive applications. However, the
> job submission and application interaction mechanisms built into Apache
> Spark are insufficient and cumbersome for multi-user interactive
> applications.
>
> The primary mechanism for applications to submit Spark jobs is via
> spark-submit
> (http://spark.apache.org/docs/latest/submitting-applications.html), which
> is
> available as a command line tool as well as a programmatic API. However,
> spark-submit has the following limitations that make it difficult to build
> interactive applications: It is slow: each invocation of spark-submit
> involves a setup phase where cluster resources are acquired, new processes
> are forked, etc. This setup phase runs for many seconds, or even minutes,
> and hence is too slow for interactive applications.  It is cumbersome and
> lacks flexibility: application code and dependencies have to be
> pre-compiled
> and submitted as jars, and can not be submitted interactively.
>
> Apache Spark comes with an ODBC/JDBC server, which can be used to submit
> SQL
> queries to Spark. However, this solution is limited to SQL and does not
> allow the client to leverage the rest of the Spark API, such as RDDs, MLlib
> and Streaming.
>
> A third way of using Spark is via its command-line shell, which allows the
> interactive submission of snippets of Spark code. However, the shell
> entails
> running Spark code on the client machine and hence is not a viable
> mechanism
> for remote clients to submit Spark jobs.
>
> Livy solves the limitations of the above three mechanisms, and provides the
> full Spark API as a multi-tenant service to remote clients.
>
> Since the open source release of Livy in late 2015, we have seen tremendous
> interest among a diverse set of application developers and ISVs that want
> to
> build applications with Apache Spark. To make Livy a robust and flexible
> solution that will enable a broad and growing set of applications, it is
> important to grow a large and varied community of contributors.
>
> = Initial Goals =
>
> Move existing codebase, website, documentation and mailing
> lists to Apache-hosted infrastructure Work with the infrastructure team to
> implement and approve our code review, build, and testing workflows in the
> context of the ASF Incremental development and releases per Apache
> guidelines
>
> = Current Status =
>
> The Livy project began at Cloudera, as a part of the Hue
> project. Cloudera soon realized the broad applicability of Livy, and
> separated it out into an independent project in Nov 2015.
>
> == Releases ==
>
> Livy has undergone two public releases, tagged here:
>
> * https://github.com/cloudera/livy/releases/tag/v0.2.0
> * https://github.com/cloudera/livy/releases/tag/v0.3.0
>
> Tarballs and zip files were created for each release and hosted on github.
> Upon joining the incubator, we will adopt a more typical ASF release
> process.
>
> == Source ==
>
> Livy’s source is currently hosted on Github at:
>
> https://github.com/cloudera/livy
>
> This repository will be transitioned to Apache’s git hosting during
> incubation.
>
> == Code review ==
>
> Livy’s code reviews are currently public and hosted on
> github as pull request reviews at: https://github.com/cloudera/livy/pulls
> The Livy developer community so far is happy with github pull request
> reviews and hopes to continue this after being admitted to the ASF.
>
> == Issue Tracking ==
>
> Livy’s bug and feature tracking is hosted on JIRA at:
> https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
> contains bugs and development discussion dating back 1 year and will
> provide
> an initial seed for the ASF JIRA
>
> == Community Discussion ==
>
> Livy has several public discussion forums:
>
> * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
> * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user
>
> == Development Practices ==
>
> The Livy project follows a review before commit philosophy. Every commit
> automatically runs through the unit tests and generates coverage reports
> presented as a pull request comment. Our experience with this process leads
> us to believe that it helps ease new contributors into the project. They
> get
> feedback quickly on common mistakes, lowering the burden on reviewers.
> Those
> same reviewers get to lead by example, showing the new contributors that we
> value feedback within our community even when changes are done by more
> experienced folks.
>
> == Meritocracy ==
>
> We believe strongly in meritocracy when electing committers and PMC
> members.
> In the past few months, the project has added two new committers from two
> different organisations, in recognition of their significant contributions
> to the project. We will encourage contributions and participation of all
> types, and ensure that contributors are appropriately recognized.
>
> == Community ==
>
> Though Livy is relatively new as a standalone open source project, it has
> already seen promising growth in its community across several
> organizations:
> Cloudera is the original development sponsor for Livy Microsoft pushed the
> development of the interpreter fixing high availability issues and adding
> additional features.  Hortonworks has contributed the security features to
> Livy allowing kerberos and impersonation to work with Spark IBM is starting
> to make contributions to the Livy project A number of other patches
> contributed by community members
>
> Livy currently relies on Google Groups for mailing lists. These lists have
> been active since the end of 2015/start of 2016. Currently, Livy’s user
> mailing list has 173 subscribers and has hosted a total of 227 topic
> threads. Livy’s developer list has 49 subscribers and has hosted 79 topic
> threads.
>
> == Core Developers ==
>
> The early contributions to Livy were made by Cloudera engineers. In 2016,
> engineers from Microsoft and Hortonworks joined the core developer
> community.
>
> == Alignment ==
>
> Livy is built upon Apache Spark, and other Apache projects like Apache
> Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
> community connections combined with our focus on development practices that
> emphasize community engagement with a path to meritocratic recognition
> naturally align us with the ASF.
>
> = Known Risks =
> == Orphaned Products ==
>
> The risk of Livy being abandoned is low because it is supported by three
> major big-data software vendors.  Moreover, Livy is already used to power
> multiple releases of services and products used in production.
>
> == Inexperience with Open Source ==
>
> Several of the initial committers are experienced open source developers,
> several being committers and/or PMC members on other ASF projects (Spark,
> YARN).
>
> == Homogenous Developers ==
>
> The project already has a diverse developer base. It has contributions from
> 3 major organisations (Cloudera, Microsoft and Hortonworks), and is used in
> diverse applications, in diverse settings (On-Prem and Cloud).
>
> == Reliance on salaried Developers ==
>
> The existing contributors to the Livy project have been made by salaried
> engineers from Cloudera, Microsoft and Hortonworks. Since there are three
> major organisations involved, the risk of reliance on a single group of
> salaried developers is mitigated. The Livy user base is diverse, with users
> from across the globe, including users from academic settings. We aim to
> further diversify the Livy user and contributor base.
>
> == Relationships with other Apache projects ==
>
> Livy is closely tied to the Apache Spark project and currently addresses
> the
> scenarios for a REST based batch and interactive gateway for Spark jobs on
> YARN. Given the growing number of integrations with Livy, keeping it
> outside
> of Apache Spark aligns with the desire of the Apache Spark community to
> reduce the number of external dependencies in the Spark project.
> Specifically, the Apache Spark community has previously expressed a desire
> to keep job servers independent from the project.<<FootNote(See, for
> example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
> Furthermore, while Livy common usage is closely tied to Spark deployments
> right now, its core building blocks can be reused elsewhere.  Livy’s Remote
> REPL could be used as a library for interactive scenarios in non-Spark
> projects. In the future, integrations with cluster managers like Apache
> Mesos and others could also be added.
>
> The features provided by Livy have already been integrated with existing
> projects like Jupyter and Apache Zeppelin for their interactive Spark use
> cases. This validates the need for a project like Livy and provides an
> active downstream user base that the Livy community can interact with to
> seed future interest in the project.
>
> Livy serves a similar purpose to Apache Toree (incubating) but differs in
> making session management, security and impersonation a focal design point.
>
> == An Excessive Fascination with the Apache Brand ==
>
> The primary motivation for submitting Livy to the ASF is to grow a diverse
> and strong community. We wish to encourage diverse organisations, including
> ISVs, to adopt Livy and contribute to Livy without any concerns about
> ownership or licensing.
>
> = Documentation =
>
> Documentation can be found on the Livy website http://livy.io/ The Livy
> web
> site is version controlled on the ‘gh-pages’ branch of the above repository
> Additional documentation is provided on the github wiki:
> https://github.com/cloudera/livy/wiki APis are documented within the
> source
> code as JavaDoc style documentation comments.
>
> = Initial Source =
>
> The initial source code for Livy is hosted at
>
> https://github.com/cloudera/livy
>
> = Source and Intellectual Property submission plan =
>
> The Livy codebase and web site is currently hosted on GitHub and will be
> transitioned to the ASF repositories during incubation. Livy is already
> licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
> CCLAs from all committers.  There are, however, some contributions recently
> from authors that have not signed the CCLA and ICLA. If necessary for a
> successful SGA, we’ll seek the necessary documentation or replace the
> contributions.
>
> The “Livy” name is not a registered trademark. We will need to do a
> trademark search and make sure it is available for the Apache Foundation
> prior to graduation.
>
> Cloudera currently owns the domain name: http://livy.io/ which will be
> transferred to the ASF and redirected to the official page during
> incubation.
>
> = External Dependencies =
>
> The list below covers the non-Apache dependencies of the project and their
> licenses.
>
>  * Jetty: Apache 2.0
>  * Dropwizard Metrics: Apache 2.0
>  * FasterXML Jackson: Apache 2.0
>  * Netty: Apache 2.0
>  * Scala: BSD
>  * Py4J: BSD
>  * Scalatra: BSD
>
> Build/test-only dependencies:
>
>  * Mockito: MIT
>  * JUnit: Eclipse
>
> = Required Resources =
> == Mailing Lists ==
>
>  * [hidden email] (PPMC)
>  * [hidden email] (dev mailing list)
>  * [hidden email] (User questions)
>  * [hidden email] (subscribers shouldn’t be able to
> post)
>  * [hidden email] (subscribers shouldn’t be able to
> post)
>
> == Git Repository ==
>
> git://git.apache.org/livy
>
> == Issue Tracking ==
>
> We would like to import our current JIRA project into the ASF JIRA, such
> that our historical commit message and code comments continue to reference
> the appropriate bug numbers.
>
> = Initial Committers =
>
>  * Marcelo Vanzin ([hidden email])
>  * Alex Man ([hidden email])
>  * Jeff Zhang ([hidden email])
>  * Saisai Shao ([hidden email])
>  * Kostas Sakellis ([hidden email])
>
> = Affiliations =
>
> The initial set of committers includes people employed by Cloudera and
> Hortonworks as well as one person currently unaffiliated with an employer.
>
> = Additional Interested Contributors =
>
> Those interested in getting involved with the project as we enter
> incubation
> are encourage to list themselves here.
>
>  * < add here >
>
> = Sponsors =
> == Champion ==
>
>  * Sean Busbey ([hidden email])
>
> == Nominated Mentors ==
>
>  * Bikas Saha ([hidden email])
>  * Brock Noland ([hidden email])
>
> == Sponsoring Entity ==
>
> We ask that the Incubator PMC sponsor this proposal.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
> --
Sent from my Mobile device
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Jitendra Pandey
+1

On 5/22/17, 1:01 AM, "Luciano Resende" <[hidden email]> wrote:

    +1
   
    Also, I see the proposal is short on mentors, so feel free to include me as
    a mentor for the project.
   
    Thanks
   
    On Fri, May 19, 2017 at 4:45 PM Sean Busbey <[hidden email]> wrote:
   
    > Dear Apache Incubator Community,
    >
    > I'm excited to present for discussion a proposal to move Livy into
    > incubation. Livy is web service that exposes a REST interface for managing
    > long running Apache Spark contexts in your cluster. With Livy, new
    > applications can be built on top of Apache Spark that require fine grained
    > interaction with many Spark contexts.
    >
    > The proposal is on the wiki at the following page as well as copied in the
    > email below:
    >
    > https://wiki.apache.org/incubator/LivyProposal
    >
    > In addition to welcoming feedback on the proposal, we are actively seeking
    > one or more additional mentors. We also have included a section for
    > interested folks to ensure they get added to the mailing lists, presuming
    > Livy gets accepted for incubation.
    >
    > ---- LivyProposal
    >
    > = Abstract =
    >
    > Livy is web service that exposes a REST interface for managing
    > long running Apache Spark contexts in your cluster. With Livy, new
    > applications can be built on top of Apache Spark that require fine grained
    > interaction with many Spark contexts.
    >
    > = Proposal =
    >
    > Livy is an open-source REST service for Apache Spark. Livy
    > enables applications to submit Spark applications and retrieve results
    > without a co-location requirement on the Spark cluster.
    >
    > We propose to contribute the Livy codebase and associated artifacts (e.g.
    > documentation, web-site context etc) to the Apache Software Foundation.
    >
    > = Background =
    >
    > Apache Spark is a fast and general purpose distributed
    > compute engine, with a versatile API. It enables processing of large
    > quantities of static data distributed over a cluster of machines, as well
    > as
    > processing of continuous streams of data. It is the preferred distributed
    > data processing engine for data engineering, stream processing and data
    > science workloads. Each Spark application uses a construct called the
    > SparkContext, which is the application’s connection or entry point to the
    > Spark engine. Each Spark application will have its own SparkContext.
    >
    > Livy enables clients to interact with one or more Spark sessions through
    > the
    > Livy Server, which acts as a proxy layer. Livy Clients have fine grained
    > control over the lifecycle of the Spark sessions, as well as the ability to
    > submit jobs and retrieve results, all over HTTP.  Clients have two modes of
    > interaction: RPC Client API, available in Java and Python, which allows
    > results to be retrieved as Java or Python objects. The serialization and
    > deserialization of the results is handled by the Livy framework.  HTTP
    > based
    > API that allows submission of code snippets, and retrieval of the results
    > in
    > different formats.
    >
    > Multi-tenant resource allocation and security: Livy enables multiple
    > independent Spark sessions to be managed simultaneously. Multiple clients
    > can also interact simultaneously with the same Spark session and share the
    > resources of that Spark session. Livy can also enforce secure,
    > authenticated
    > communication between the clients and their respective Spark sessions.
    >
    > More information on Livy can be found at the existing open source website:
    > http://livy.io/
    >
    > = Rationale =
    >
    > Users want to use Spark’s powerful processing engine and API
    > as the data processing backend for interactive applications. However, the
    > job submission and application interaction mechanisms built into Apache
    > Spark are insufficient and cumbersome for multi-user interactive
    > applications.
    >
    > The primary mechanism for applications to submit Spark jobs is via
    > spark-submit
    > (http://spark.apache.org/docs/latest/submitting-applications.html), which
    > is
    > available as a command line tool as well as a programmatic API. However,
    > spark-submit has the following limitations that make it difficult to build
    > interactive applications: It is slow: each invocation of spark-submit
    > involves a setup phase where cluster resources are acquired, new processes
    > are forked, etc. This setup phase runs for many seconds, or even minutes,
    > and hence is too slow for interactive applications.  It is cumbersome and
    > lacks flexibility: application code and dependencies have to be
    > pre-compiled
    > and submitted as jars, and can not be submitted interactively.
    >
    > Apache Spark comes with an ODBC/JDBC server, which can be used to submit
    > SQL
    > queries to Spark. However, this solution is limited to SQL and does not
    > allow the client to leverage the rest of the Spark API, such as RDDs, MLlib
    > and Streaming.
    >
    > A third way of using Spark is via its command-line shell, which allows the
    > interactive submission of snippets of Spark code. However, the shell
    > entails
    > running Spark code on the client machine and hence is not a viable
    > mechanism
    > for remote clients to submit Spark jobs.
    >
    > Livy solves the limitations of the above three mechanisms, and provides the
    > full Spark API as a multi-tenant service to remote clients.
    >
    > Since the open source release of Livy in late 2015, we have seen tremendous
    > interest among a diverse set of application developers and ISVs that want
    > to
    > build applications with Apache Spark. To make Livy a robust and flexible
    > solution that will enable a broad and growing set of applications, it is
    > important to grow a large and varied community of contributors.
    >
    > = Initial Goals =
    >
    > Move existing codebase, website, documentation and mailing
    > lists to Apache-hosted infrastructure Work with the infrastructure team to
    > implement and approve our code review, build, and testing workflows in the
    > context of the ASF Incremental development and releases per Apache
    > guidelines
    >
    > = Current Status =
    >
    > The Livy project began at Cloudera, as a part of the Hue
    > project. Cloudera soon realized the broad applicability of Livy, and
    > separated it out into an independent project in Nov 2015.
    >
    > == Releases ==
    >
    > Livy has undergone two public releases, tagged here:
    >
    > * https://github.com/cloudera/livy/releases/tag/v0.2.0
    > * https://github.com/cloudera/livy/releases/tag/v0.3.0
    >
    > Tarballs and zip files were created for each release and hosted on github.
    > Upon joining the incubator, we will adopt a more typical ASF release
    > process.
    >
    > == Source ==
    >
    > Livy’s source is currently hosted on Github at:
    >
    > https://github.com/cloudera/livy
    >
    > This repository will be transitioned to Apache’s git hosting during
    > incubation.
    >
    > == Code review ==
    >
    > Livy’s code reviews are currently public and hosted on
    > github as pull request reviews at: https://github.com/cloudera/livy/pulls
    > The Livy developer community so far is happy with github pull request
    > reviews and hopes to continue this after being admitted to the ASF.
    >
    > == Issue Tracking ==
    >
    > Livy’s bug and feature tracking is hosted on JIRA at:
    > https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
    > contains bugs and development discussion dating back 1 year and will
    > provide
    > an initial seed for the ASF JIRA
    >
    > == Community Discussion ==
    >
    > Livy has several public discussion forums:
    >
    > * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
    > * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user
    >
    > == Development Practices ==
    >
    > The Livy project follows a review before commit philosophy. Every commit
    > automatically runs through the unit tests and generates coverage reports
    > presented as a pull request comment. Our experience with this process leads
    > us to believe that it helps ease new contributors into the project. They
    > get
    > feedback quickly on common mistakes, lowering the burden on reviewers.
    > Those
    > same reviewers get to lead by example, showing the new contributors that we
    > value feedback within our community even when changes are done by more
    > experienced folks.
    >
    > == Meritocracy ==
    >
    > We believe strongly in meritocracy when electing committers and PMC
    > members.
    > In the past few months, the project has added two new committers from two
    > different organisations, in recognition of their significant contributions
    > to the project. We will encourage contributions and participation of all
    > types, and ensure that contributors are appropriately recognized.
    >
    > == Community ==
    >
    > Though Livy is relatively new as a standalone open source project, it has
    > already seen promising growth in its community across several
    > organizations:
    > Cloudera is the original development sponsor for Livy Microsoft pushed the
    > development of the interpreter fixing high availability issues and adding
    > additional features.  Hortonworks has contributed the security features to
    > Livy allowing kerberos and impersonation to work with Spark IBM is starting
    > to make contributions to the Livy project A number of other patches
    > contributed by community members
    >
    > Livy currently relies on Google Groups for mailing lists. These lists have
    > been active since the end of 2015/start of 2016. Currently, Livy’s user
    > mailing list has 173 subscribers and has hosted a total of 227 topic
    > threads. Livy’s developer list has 49 subscribers and has hosted 79 topic
    > threads.
    >
    > == Core Developers ==
    >
    > The early contributions to Livy were made by Cloudera engineers. In 2016,
    > engineers from Microsoft and Hortonworks joined the core developer
    > community.
    >
    > == Alignment ==
    >
    > Livy is built upon Apache Spark, and other Apache projects like Apache
    > Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
    > community connections combined with our focus on development practices that
    > emphasize community engagement with a path to meritocratic recognition
    > naturally align us with the ASF.
    >
    > = Known Risks =
    > == Orphaned Products ==
    >
    > The risk of Livy being abandoned is low because it is supported by three
    > major big-data software vendors.  Moreover, Livy is already used to power
    > multiple releases of services and products used in production.
    >
    > == Inexperience with Open Source ==
    >
    > Several of the initial committers are experienced open source developers,
    > several being committers and/or PMC members on other ASF projects (Spark,
    > YARN).
    >
    > == Homogenous Developers ==
    >
    > The project already has a diverse developer base. It has contributions from
    > 3 major organisations (Cloudera, Microsoft and Hortonworks), and is used in
    > diverse applications, in diverse settings (On-Prem and Cloud).
    >
    > == Reliance on salaried Developers ==
    >
    > The existing contributors to the Livy project have been made by salaried
    > engineers from Cloudera, Microsoft and Hortonworks. Since there are three
    > major organisations involved, the risk of reliance on a single group of
    > salaried developers is mitigated. The Livy user base is diverse, with users
    > from across the globe, including users from academic settings. We aim to
    > further diversify the Livy user and contributor base.
    >
    > == Relationships with other Apache projects ==
    >
    > Livy is closely tied to the Apache Spark project and currently addresses
    > the
    > scenarios for a REST based batch and interactive gateway for Spark jobs on
    > YARN. Given the growing number of integrations with Livy, keeping it
    > outside
    > of Apache Spark aligns with the desire of the Apache Spark community to
    > reduce the number of external dependencies in the Spark project.
    > Specifically, the Apache Spark community has previously expressed a desire
    > to keep job servers independent from the project.<<FootNote(See, for
    > example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
    > Furthermore, while Livy common usage is closely tied to Spark deployments
    > right now, its core building blocks can be reused elsewhere.  Livy’s Remote
    > REPL could be used as a library for interactive scenarios in non-Spark
    > projects. In the future, integrations with cluster managers like Apache
    > Mesos and others could also be added.
    >
    > The features provided by Livy have already been integrated with existing
    > projects like Jupyter and Apache Zeppelin for their interactive Spark use
    > cases. This validates the need for a project like Livy and provides an
    > active downstream user base that the Livy community can interact with to
    > seed future interest in the project.
    >
    > Livy serves a similar purpose to Apache Toree (incubating) but differs in
    > making session management, security and impersonation a focal design point.
    >
    > == An Excessive Fascination with the Apache Brand ==
    >
    > The primary motivation for submitting Livy to the ASF is to grow a diverse
    > and strong community. We wish to encourage diverse organisations, including
    > ISVs, to adopt Livy and contribute to Livy without any concerns about
    > ownership or licensing.
    >
    > = Documentation =
    >
    > Documentation can be found on the Livy website http://livy.io/ The Livy
    > web
    > site is version controlled on the ‘gh-pages’ branch of the above repository
    > Additional documentation is provided on the github wiki:
    > https://github.com/cloudera/livy/wiki APis are documented within the
    > source
    > code as JavaDoc style documentation comments.
    >
    > = Initial Source =
    >
    > The initial source code for Livy is hosted at
    >
    > https://github.com/cloudera/livy
    >
    > = Source and Intellectual Property submission plan =
    >
    > The Livy codebase and web site is currently hosted on GitHub and will be
    > transitioned to the ASF repositories during incubation. Livy is already
    > licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
    > CCLAs from all committers.  There are, however, some contributions recently
    > from authors that have not signed the CCLA and ICLA. If necessary for a
    > successful SGA, we’ll seek the necessary documentation or replace the
    > contributions.
    >
    > The “Livy” name is not a registered trademark. We will need to do a
    > trademark search and make sure it is available for the Apache Foundation
    > prior to graduation.
    >
    > Cloudera currently owns the domain name: http://livy.io/ which will be
    > transferred to the ASF and redirected to the official page during
    > incubation.
    >
    > = External Dependencies =
    >
    > The list below covers the non-Apache dependencies of the project and their
    > licenses.
    >
    >  * Jetty: Apache 2.0
    >  * Dropwizard Metrics: Apache 2.0
    >  * FasterXML Jackson: Apache 2.0
    >  * Netty: Apache 2.0
    >  * Scala: BSD
    >  * Py4J: BSD
    >  * Scalatra: BSD
    >
    > Build/test-only dependencies:
    >
    >  * Mockito: MIT
    >  * JUnit: Eclipse
    >
    > = Required Resources =
    > == Mailing Lists ==
    >
    >  * [hidden email] (PPMC)
    >  * [hidden email] (dev mailing list)
    >  * [hidden email] (User questions)
    >  * [hidden email] (subscribers shouldn’t be able to
    > post)
    >  * [hidden email] (subscribers shouldn’t be able to
    > post)
    >
    > == Git Repository ==
    >
    > git://git.apache.org/livy
    >
    > == Issue Tracking ==
    >
    > We would like to import our current JIRA project into the ASF JIRA, such
    > that our historical commit message and code comments continue to reference
    > the appropriate bug numbers.
    >
    > = Initial Committers =
    >
    >  * Marcelo Vanzin ([hidden email])
    >  * Alex Man ([hidden email])
    >  * Jeff Zhang ([hidden email])
    >  * Saisai Shao ([hidden email])
    >  * Kostas Sakellis ([hidden email])
    >
    > = Affiliations =
    >
    > The initial set of committers includes people employed by Cloudera and
    > Hortonworks as well as one person currently unaffiliated with an employer.
    >
    > = Additional Interested Contributors =
    >
    > Those interested in getting involved with the project as we enter
    > incubation
    > are encourage to list themselves here.
    >
    >  * < add here >
    >
    > = Sponsors =
    > == Champion ==
    >
    >  * Sean Busbey ([hidden email])
    >
    > == Nominated Mentors ==
    >
    >  * Bikas Saha ([hidden email])
    >  * Brock Noland ([hidden email])
    >
    > == Sponsoring Entity ==
    >
    > We ask that the Incubator PMC sponsor this proposal.
    >
    >
    > ---------------------------------------------------------------------
    > To unsubscribe, e-mail: [hidden email]
    > For additional commands, e-mail: [hidden email]
    >
    > --
    Sent from my Mobile device
   


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Sean Busbey-2
In reply to this post by Sean Busbey-2
On 2017-05-22 03:01 (-0500), Luciano Resende <[hidden email]> wrote:
>  1>
>
> Also, I see the proposal is short on mentors, so feel free to include me as>
> a mentor for the project.>
>
> Thanks>
>


Thanks Luciano! I've added you to the wiki page as a mentor.

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Sean Busbey-2
In reply to this post by Sean Busbey-2
On 2017-05-21 09:46 (-0500), "John D. Ament" <[hidden email]> wrote:

>
> On Fri, May 19, 2017 at 7:45 PM Sean Busbey <[hidden email]> wrote:
> >
> > == Reliance on salaried Developers ==
> >
> > The existing contributors to the Livy project have been made by salaried
> > engineers from Cloudera, Microsoft and Hortonworks. Since there are three
> > major organisations involved, the risk of reliance on a single group of
> > salaried developers is mitigated. The Livy user base is diverse, with users
> > from across the globe, including users from academic settings. We aim to
> > further diversify the Livy user and contributor base.
> >
>
> There's a disconnect between this paragraph and the initial committers
> list. Specifically, no one from Microsoft is represented (as best as I can
> tell).

Ah, this lack of clarity is my fault as an editor. One of the initial
committers was recently employed by Microsoft but is now in the
process of changing employers. Another person formally affiliated with
both the project and Microsoft has decided not to continue
participating.

I could rephrase this to talk about the contributions made to date as
being from individuals then in the employ of three major companies.
Then call out the initial committer list as from two of those and one
unaffiliated. Would that read clearer?

> > Cloudera currently owns the domain name: http://livy.io/ which will be
> > transferred to the ASF and redirected to the official page during
> > incubation.
> >
> >
>
> I'm assuming that the incoming project is aware that we expect the main dev
> landing page to be livy.incubator.apache.org . We will want to track this
> as a project specific item.


Yep, once all the docs are moved over to ASF infrastructure we can
just have the current domain act as a redirect. Should I call this out
in the proposal?


> > == Git Repository ==
> >
> > git://git.apache.org/livy
> >
>
> Just to confirm - it'll be incubator-livy, not just livy.

right right. I'll correct this when making the other edits.


> > == Issue Tracking ==
> >
> > We would like to import our current JIRA project into the ASF JIRA, such
> > that our historical commit message and code comments continue to reference
> > the appropriate bug numbers.
> >
>
> I would recommend reaching out to infra to see if the import is possible
> before voting on the project. Otherwise you'll need to list out an la

Sure I can reach out. I've seen this done a few times, so I consider
it low risk. The end of your line appears to have been lost, what's
the "Otherwise..." ?

> > = Sponsors =
> > == Champion ==
> >
> > * Sean Busbey ([hidden email])
> >
> > == Nominated Mentors ==
> >
> > * Bikas Saha ([hidden email])
> > * Brock Noland ([hidden email])
>
>
> A couple of points:
>
> - Sean, while the champion and mentor roles are separate, we do hope that
> all champions will continue on as a mentor. If this is your intention
> please add yourself.

After having to withdraw from mentoring a couple of podlings at the
end of last year I am conservative about what my volunteer time looks
like right now. I'm certain I can spare the time to help the Livy
community get introduced to the incubator. I'm not certain beyond
that, so I am not listed as a formal mentor.

>  - All mentors must be on the IPMC. Foundation membership isn't a
> requirement, however most people use membership to get access to the IPMC.
> If Bikas wants to be a mentor, he'll need to join the IPMC otherwise you'll
> need to find 2 mentors.

I was pretty sure Bikas had already done this step. I'll chase this
down and find the disconnect.

> - Do the proposed mentors have a relationship to the incoming project, e.g.
> do they care if it succeeds from a corporate interest standpoint?

I'll let the mentors answer for themselves here, as I won't presume to
know their specific motivations for volunteering.

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

John D. Ament-2
On Mon, May 22, 2017 at 3:30 PM Sean Busbey <[hidden email]> wrote:

> On 2017-05-21 09:46 (-0500), "John D. Ament" <[hidden email]>
> wrote:
> >
> > On Fri, May 19, 2017 at 7:45 PM Sean Busbey <[hidden email]> wrote:
> > >
> > > == Reliance on salaried Developers ==
> > >
> > > The existing contributors to the Livy project have been made by
> salaried
> > > engineers from Cloudera, Microsoft and Hortonworks. Since there are
> three
> > > major organisations involved, the risk of reliance on a single group of
> > > salaried developers is mitigated. The Livy user base is diverse, with
> users
> > > from across the globe, including users from academic settings. We aim
> to
> > > further diversify the Livy user and contributor base.
> > >
> >
> > There's a disconnect between this paragraph and the initial committers
> > list. Specifically, no one from Microsoft is represented (as best as I
> can
> > tell).
>
> Ah, this lack of clarity is my fault as an editor. One of the initial
> committers was recently employed by Microsoft but is now in the
> process of changing employers. Another person formally affiliated with
> both the project and Microsoft has decided not to continue
> participating.
>
> I could rephrase this to talk about the contributions made to date as
> being from individuals then in the employ of three major companies.
> Then call out the initial committer list as from two of those and one
> unaffiliated. Would that read clearer?
>
>
I think the explanation is enough, nothing to change.


> > > Cloudera currently owns the domain name: http://livy.io/ which will be
> > > transferred to the ASF and redirected to the official page during
> > > incubation.
> > >
> > >
> >
> > I'm assuming that the incoming project is aware that we expect the main
> dev
> > landing page to be livy.incubator.apache.org . We will want to track
> this
> > as a project specific item.
>
>
> Yep, once all the docs are moved over to ASF infrastructure we can
> just have the current domain act as a redirect. Should I call this out
> in the proposal?
>
>
It would be good to list out the goal of moving to livy.i.a.o but not
needed.


>
> > > == Git Repository ==
> > >
> > > git://git.apache.org/livy
> > >
> >
> > Just to confirm - it'll be incubator-livy, not just livy.
>
> right right. I'll correct this when making the other edits.
>
>
> > > == Issue Tracking ==
> > >
> > > We would like to import our current JIRA project into the ASF JIRA,
> such
> > > that our historical commit message and code comments continue to
> reference
> > > the appropriate bug numbers.
> > >
> >
> > I would recommend reaching out to infra to see if the import is possible
> > before voting on the project. Otherwise you'll need to list out an la
>
> Sure I can reach out. I've seen this done a few times, so I consider
> it low risk. The end of your line appears to have been lost, what's
> the "Otherwise..." ?
>
>
Yeah, I'm not sure what happened there either.  I checked my mail client,
E_NOCLUE.

"list out an alternative" is what I was typing.


> > > = Sponsors =
> > > == Champion ==
> > >
> > > * Sean Busbey ([hidden email])
> > >
> > > == Nominated Mentors ==
> > >
> > > * Bikas Saha ([hidden email])
> > > * Brock Noland ([hidden email])
> >
> >
> > A couple of points:
> >
> > - Sean, while the champion and mentor roles are separate, we do hope that
> > all champions will continue on as a mentor. If this is your intention
> > please add yourself.
>
> After having to withdraw from mentoring a couple of podlings at the
> end of last year I am conservative about what my volunteer time looks
> like right now. I'm certain I can spare the time to help the Livy
> community get introduced to the incubator. I'm not certain beyond
> that, so I am not listed as a formal mentor.
>
> >  - All mentors must be on the IPMC. Foundation membership isn't a
> > requirement, however most people use membership to get access to the
> IPMC.
> > If Bikas wants to be a mentor, he'll need to join the IPMC otherwise
> you'll
> > need to find 2 mentors.
>
> I was pretty sure Bikas had already done this step. I'll chase this
> down and find the disconnect.
>
> > - Do the proposed mentors have a relationship to the incoming project,
> e.g.
> > do they care if it succeeds from a corporate interest standpoint?
>
> I'll let the mentors answer for themselves here, as I won't presume to
> know their specific motivations for volunteering.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Kostas Sakellis
In reply to this post by Luciano Resende
On Mon, May 22, 2017 at 1:01 AM, Luciano Resende <[hidden email]>
wrote:

> +1
>
> Also, I see the proposal is short on mentors, so feel free to include me as
> a mentor for the project.
>
> Thanks
>

Thanks Luciano! Welcome onboard.


> On Fri, May 19, 2017 at 4:45 PM Sean Busbey <[hidden email]> wrote:
>
> > Dear Apache Incubator Community,
> >
> > I'm excited to present for discussion a proposal to move Livy into
> > incubation. Livy is web service that exposes a REST interface for
> managing
> > long running Apache Spark contexts in your cluster. With Livy, new
> > applications can be built on top of Apache Spark that require fine
> grained
> > interaction with many Spark contexts.
> >
> > The proposal is on the wiki at the following page as well as copied in
> the
> > email below:
> >
> > https://wiki.apache.org/incubator/LivyProposal
> >
> > In addition to welcoming feedback on the proposal, we are actively
> seeking
> > one or more additional mentors. We also have included a section for
> > interested folks to ensure they get added to the mailing lists, presuming
> > Livy gets accepted for incubation.
> >
> > ---- LivyProposal
> >
> > = Abstract =
> >
> > Livy is web service that exposes a REST interface for managing
> > long running Apache Spark contexts in your cluster. With Livy, new
> > applications can be built on top of Apache Spark that require fine
> grained
> > interaction with many Spark contexts.
> >
> > = Proposal =
> >
> > Livy is an open-source REST service for Apache Spark. Livy
> > enables applications to submit Spark applications and retrieve results
> > without a co-location requirement on the Spark cluster.
> >
> > We propose to contribute the Livy codebase and associated artifacts (e.g.
> > documentation, web-site context etc) to the Apache Software Foundation.
> >
> > = Background =
> >
> > Apache Spark is a fast and general purpose distributed
> > compute engine, with a versatile API. It enables processing of large
> > quantities of static data distributed over a cluster of machines, as well
> > as
> > processing of continuous streams of data. It is the preferred distributed
> > data processing engine for data engineering, stream processing and data
> > science workloads. Each Spark application uses a construct called the
> > SparkContext, which is the application’s connection or entry point to the
> > Spark engine. Each Spark application will have its own SparkContext.
> >
> > Livy enables clients to interact with one or more Spark sessions through
> > the
> > Livy Server, which acts as a proxy layer. Livy Clients have fine grained
> > control over the lifecycle of the Spark sessions, as well as the ability
> to
> > submit jobs and retrieve results, all over HTTP.  Clients have two modes
> of
> > interaction: RPC Client API, available in Java and Python, which allows
> > results to be retrieved as Java or Python objects. The serialization and
> > deserialization of the results is handled by the Livy framework.  HTTP
> > based
> > API that allows submission of code snippets, and retrieval of the results
> > in
> > different formats.
> >
> > Multi-tenant resource allocation and security: Livy enables multiple
> > independent Spark sessions to be managed simultaneously. Multiple clients
> > can also interact simultaneously with the same Spark session and share
> the
> > resources of that Spark session. Livy can also enforce secure,
> > authenticated
> > communication between the clients and their respective Spark sessions.
> >
> > More information on Livy can be found at the existing open source
> website:
> > http://livy.io/
> >
> > = Rationale =
> >
> > Users want to use Spark’s powerful processing engine and API
> > as the data processing backend for interactive applications. However, the
> > job submission and application interaction mechanisms built into Apache
> > Spark are insufficient and cumbersome for multi-user interactive
> > applications.
> >
> > The primary mechanism for applications to submit Spark jobs is via
> > spark-submit
> > (http://spark.apache.org/docs/latest/submitting-applications.html),
> which
> > is
> > available as a command line tool as well as a programmatic API. However,
> > spark-submit has the following limitations that make it difficult to
> build
> > interactive applications: It is slow: each invocation of spark-submit
> > involves a setup phase where cluster resources are acquired, new
> processes
> > are forked, etc. This setup phase runs for many seconds, or even minutes,
> > and hence is too slow for interactive applications.  It is cumbersome and
> > lacks flexibility: application code and dependencies have to be
> > pre-compiled
> > and submitted as jars, and can not be submitted interactively.
> >
> > Apache Spark comes with an ODBC/JDBC server, which can be used to submit
> > SQL
> > queries to Spark. However, this solution is limited to SQL and does not
> > allow the client to leverage the rest of the Spark API, such as RDDs,
> MLlib
> > and Streaming.
> >
> > A third way of using Spark is via its command-line shell, which allows
> the
> > interactive submission of snippets of Spark code. However, the shell
> > entails
> > running Spark code on the client machine and hence is not a viable
> > mechanism
> > for remote clients to submit Spark jobs.
> >
> > Livy solves the limitations of the above three mechanisms, and provides
> the
> > full Spark API as a multi-tenant service to remote clients.
> >
> > Since the open source release of Livy in late 2015, we have seen
> tremendous
> > interest among a diverse set of application developers and ISVs that want
> > to
> > build applications with Apache Spark. To make Livy a robust and flexible
> > solution that will enable a broad and growing set of applications, it is
> > important to grow a large and varied community of contributors.
> >
> > = Initial Goals =
> >
> > Move existing codebase, website, documentation and mailing
> > lists to Apache-hosted infrastructure Work with the infrastructure team
> to
> > implement and approve our code review, build, and testing workflows in
> the
> > context of the ASF Incremental development and releases per Apache
> > guidelines
> >
> > = Current Status =
> >
> > The Livy project began at Cloudera, as a part of the Hue
> > project. Cloudera soon realized the broad applicability of Livy, and
> > separated it out into an independent project in Nov 2015.
> >
> > == Releases ==
> >
> > Livy has undergone two public releases, tagged here:
> >
> > * https://github.com/cloudera/livy/releases/tag/v0.2.0
> > * https://github.com/cloudera/livy/releases/tag/v0.3.0
> >
> > Tarballs and zip files were created for each release and hosted on
> github.
> > Upon joining the incubator, we will adopt a more typical ASF release
> > process.
> >
> > == Source ==
> >
> > Livy’s source is currently hosted on Github at:
> >
> > https://github.com/cloudera/livy
> >
> > This repository will be transitioned to Apache’s git hosting during
> > incubation.
> >
> > == Code review ==
> >
> > Livy’s code reviews are currently public and hosted on
> > github as pull request reviews at: https://github.com/cloudera/
> livy/pulls
> > The Livy developer community so far is happy with github pull request
> > reviews and hopes to continue this after being admitted to the ASF.
> >
> > == Issue Tracking ==
> >
> > Livy’s bug and feature tracking is hosted on JIRA at:
> > https://issues.cloudera.org/projects/LIVY/summary This JIRA instance
> > contains bugs and development discussion dating back 1 year and will
> > provide
> > an initial seed for the ASF JIRA
> >
> > == Community Discussion ==
> >
> > Livy has several public discussion forums:
> >
> > * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-dev
> > * https://groups.google.com/a/cloudera.org/forum/#!forum/livy-user
> >
> > == Development Practices ==
> >
> > The Livy project follows a review before commit philosophy. Every commit
> > automatically runs through the unit tests and generates coverage reports
> > presented as a pull request comment. Our experience with this process
> leads
> > us to believe that it helps ease new contributors into the project. They
> > get
> > feedback quickly on common mistakes, lowering the burden on reviewers.
> > Those
> > same reviewers get to lead by example, showing the new contributors that
> we
> > value feedback within our community even when changes are done by more
> > experienced folks.
> >
> > == Meritocracy ==
> >
> > We believe strongly in meritocracy when electing committers and PMC
> > members.
> > In the past few months, the project has added two new committers from two
> > different organisations, in recognition of their significant
> contributions
> > to the project. We will encourage contributions and participation of all
> > types, and ensure that contributors are appropriately recognized.
> >
> > == Community ==
> >
> > Though Livy is relatively new as a standalone open source project, it has
> > already seen promising growth in its community across several
> > organizations:
> > Cloudera is the original development sponsor for Livy Microsoft pushed
> the
> > development of the interpreter fixing high availability issues and adding
> > additional features.  Hortonworks has contributed the security features
> to
> > Livy allowing kerberos and impersonation to work with Spark IBM is
> starting
> > to make contributions to the Livy project A number of other patches
> > contributed by community members
> >
> > Livy currently relies on Google Groups for mailing lists. These lists
> have
> > been active since the end of 2015/start of 2016. Currently, Livy’s user
> > mailing list has 173 subscribers and has hosted a total of 227 topic
> > threads. Livy’s developer list has 49 subscribers and has hosted 79 topic
> > threads.
> >
> > == Core Developers ==
> >
> > The early contributions to Livy were made by Cloudera engineers. In 2016,
> > engineers from Microsoft and Hortonworks joined the core developer
> > community.
> >
> > == Alignment ==
> >
> > Livy is built upon Apache Spark, and other Apache projects like Apache
> > Hadoop YARN. It’s used as a building block by Apache Zeppelin.  These
> > community connections combined with our focus on development practices
> that
> > emphasize community engagement with a path to meritocratic recognition
> > naturally align us with the ASF.
> >
> > = Known Risks =
> > == Orphaned Products ==
> >
> > The risk of Livy being abandoned is low because it is supported by three
> > major big-data software vendors.  Moreover, Livy is already used to power
> > multiple releases of services and products used in production.
> >
> > == Inexperience with Open Source ==
> >
> > Several of the initial committers are experienced open source developers,
> > several being committers and/or PMC members on other ASF projects (Spark,
> > YARN).
> >
> > == Homogenous Developers ==
> >
> > The project already has a diverse developer base. It has contributions
> from
> > 3 major organisations (Cloudera, Microsoft and Hortonworks), and is used
> in
> > diverse applications, in diverse settings (On-Prem and Cloud).
> >
> > == Reliance on salaried Developers ==
> >
> > The existing contributors to the Livy project have been made by salaried
> > engineers from Cloudera, Microsoft and Hortonworks. Since there are three
> > major organisations involved, the risk of reliance on a single group of
> > salaried developers is mitigated. The Livy user base is diverse, with
> users
> > from across the globe, including users from academic settings. We aim to
> > further diversify the Livy user and contributor base.
> >
> > == Relationships with other Apache projects ==
> >
> > Livy is closely tied to the Apache Spark project and currently addresses
> > the
> > scenarios for a REST based batch and interactive gateway for Spark jobs
> on
> > YARN. Given the growing number of integrations with Livy, keeping it
> > outside
> > of Apache Spark aligns with the desire of the Apache Spark community to
> > reduce the number of external dependencies in the Spark project.
> > Specifically, the Apache Spark community has previously expressed a
> desire
> > to keep job servers independent from the project.<<FootNote(See, for
> > example, discussion of the Ooyala Spark Job Server in SPARK-818)>>
> > Furthermore, while Livy common usage is closely tied to Spark deployments
> > right now, its core building blocks can be reused elsewhere.  Livy’s
> Remote
> > REPL could be used as a library for interactive scenarios in non-Spark
> > projects. In the future, integrations with cluster managers like Apache
> > Mesos and others could also be added.
> >
> > The features provided by Livy have already been integrated with existing
> > projects like Jupyter and Apache Zeppelin for their interactive Spark use
> > cases. This validates the need for a project like Livy and provides an
> > active downstream user base that the Livy community can interact with to
> > seed future interest in the project.
> >
> > Livy serves a similar purpose to Apache Toree (incubating) but differs in
> > making session management, security and impersonation a focal design
> point.
> >
> > == An Excessive Fascination with the Apache Brand ==
> >
> > The primary motivation for submitting Livy to the ASF is to grow a
> diverse
> > and strong community. We wish to encourage diverse organisations,
> including
> > ISVs, to adopt Livy and contribute to Livy without any concerns about
> > ownership or licensing.
> >
> > = Documentation =
> >
> > Documentation can be found on the Livy website http://livy.io/ The Livy
> > web
> > site is version controlled on the ‘gh-pages’ branch of the above
> repository
> > Additional documentation is provided on the github wiki:
> > https://github.com/cloudera/livy/wiki APis are documented within the
> > source
> > code as JavaDoc style documentation comments.
> >
> > = Initial Source =
> >
> > The initial source code for Livy is hosted at
> >
> > https://github.com/cloudera/livy
> >
> > = Source and Intellectual Property submission plan =
> >
> > The Livy codebase and web site is currently hosted on GitHub and will be
> > transitioned to the ASF repositories during incubation. Livy is already
> > licensed under the Apache 2.0 license. Cloudera has collected ICLAs and
> > CCLAs from all committers.  There are, however, some contributions
> recently
> > from authors that have not signed the CCLA and ICLA. If necessary for a
> > successful SGA, we’ll seek the necessary documentation or replace the
> > contributions.
> >
> > The “Livy” name is not a registered trademark. We will need to do a
> > trademark search and make sure it is available for the Apache Foundation
> > prior to graduation.
> >
> > Cloudera currently owns the domain name: http://livy.io/ which will be
> > transferred to the ASF and redirected to the official page during
> > incubation.
> >
> > = External Dependencies =
> >
> > The list below covers the non-Apache dependencies of the project and
> their
> > licenses.
> >
> >  * Jetty: Apache 2.0
> >  * Dropwizard Metrics: Apache 2.0
> >  * FasterXML Jackson: Apache 2.0
> >  * Netty: Apache 2.0
> >  * Scala: BSD
> >  * Py4J: BSD
> >  * Scalatra: BSD
> >
> > Build/test-only dependencies:
> >
> >  * Mockito: MIT
> >  * JUnit: Eclipse
> >
> > = Required Resources =
> > == Mailing Lists ==
> >
> >  * [hidden email] (PPMC)
> >  * [hidden email] (dev mailing list)
> >  * [hidden email] (User questions)
> >  * [hidden email] (subscribers shouldn’t be able to
> > post)
> >  * [hidden email] (subscribers shouldn’t be able to
> > post)
> >
> > == Git Repository ==
> >
> > git://git.apache.org/livy
> >
> > == Issue Tracking ==
> >
> > We would like to import our current JIRA project into the ASF JIRA, such
> > that our historical commit message and code comments continue to
> reference
> > the appropriate bug numbers.
> >
> > = Initial Committers =
> >
> >  * Marcelo Vanzin ([hidden email])
> >  * Alex Man ([hidden email])
> >  * Jeff Zhang ([hidden email])
> >  * Saisai Shao ([hidden email])
> >  * Kostas Sakellis ([hidden email])
> >
> > = Affiliations =
> >
> > The initial set of committers includes people employed by Cloudera and
> > Hortonworks as well as one person currently unaffiliated with an
> employer.
> >
> > = Additional Interested Contributors =
> >
> > Those interested in getting involved with the project as we enter
> > incubation
> > are encourage to list themselves here.
> >
> >  * < add here >
> >
> > = Sponsors =
> > == Champion ==
> >
> >  * Sean Busbey ([hidden email])
> >
> > == Nominated Mentors ==
> >
> >  * Bikas Saha ([hidden email])
> >  * Brock Noland ([hidden email])
> >
> > == Sponsoring Entity ==
> >
> > We ask that the Incubator PMC sponsor this proposal.
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [hidden email]
> > For additional commands, e-mail: [hidden email]
> >
> > --
> Sent from my Mobile device
>
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Kostas Sakellis
In reply to this post by Sean Busbey-2
Thank you John for the feedback.

On Mon, May 22, 2017 at 12:30 PM, Sean Busbey <[hidden email]> wrote:

> On 2017-05-21 09:46 (-0500), "John D. Ament" <[hidden email]>
> wrote:
> >
> > On Fri, May 19, 2017 at 7:45 PM Sean Busbey <[hidden email]> wrote:
> > >
> > > == Reliance on salaried Developers ==
> > >
> > > The existing contributors to the Livy project have been made by
> salaried
> > > engineers from Cloudera, Microsoft and Hortonworks. Since there are
> three
> > > major organisations involved, the risk of reliance on a single group of
> > > salaried developers is mitigated. The Livy user base is diverse, with
> users
> > > from across the globe, including users from academic settings. We aim
> to
> > > further diversify the Livy user and contributor base.
> > >
> >
> > There's a disconnect between this paragraph and the initial committers
> > list. Specifically, no one from Microsoft is represented (as best as I
> can
> > tell).
>
> Ah, this lack of clarity is my fault as an editor. One of the initial
> committers was recently employed by Microsoft but is now in the
> process of changing employers. Another person formally affiliated with
> both the project and Microsoft has decided not to continue
> participating.
>
> I could rephrase this to talk about the contributions made to date as
> being from individuals then in the employ of three major companies.
> Then call out the initial committer list as from two of those and one
> unaffiliated. Would that read clearer?
>

I modified this section in the proposal to read: "The contributions to the
Livy project to date have been made by salaried engineers from Cloudera,
Microsoft and Hortonworks. One of the individuals on the initial committer
list has since left Microsoft and is currently unaffiliated. The remaining
contributors are from Cloudera and Hortonworks. Since there are at least
two major organizations involved, the risk of reliance on a single group of
salaried developers is mitigated. The Livy user base is diverse, with users
from across the globe, including users from academic settings. We aim to
further diversify the Livy user and contributor base."


>
> > > Cloudera currently owns the domain name: http://livy.io/ which will be
> > > transferred to the ASF and redirected to the official page during
> > > incubation.
> > >
> > >
> >
> > I'm assuming that the incoming project is aware that we expect the main
> dev
> > landing page to be livy.incubator.apache.org . We will want to track
> this
> > as a project specific item.
>
>
> Yep, once all the docs are moved over to ASF infrastructure we can
> just have the current domain act as a redirect. Should I call this out
> in the proposal?
>
> I modified the proposal to say: "Cloudera currently owns the domain name:
http://livy.io/. Once all the documentation has moved over to ASF
infrastructure, the main landing page will become livy.incubator.apache.org
and the old domain will just act as a redirect."


>
> > > == Git Repository ==
> > >
> > > git://git.apache.org/livy
> > >
> >
> > Just to confirm - it'll be incubator-livy, not just livy.
>
> right right. I'll correct this when making the other edits.
>
> Added the correction to the proposal


>
> > > == Issue Tracking ==
> > >
> > > We would like to import our current JIRA project into the ASF JIRA,
> such
> > > that our historical commit message and code comments continue to
> reference
> > > the appropriate bug numbers.
> > >
> >
> > I would recommend reaching out to infra to see if the import is possible
> > before voting on the project. Otherwise you'll need to list out an la
>
> Sure I can reach out. I've seen this done a few times, so I consider
> it low risk. The end of your line appears to have been lost, what's
> the "Otherwise..." ?
>
> > > = Sponsors =
> > > == Champion ==
> > >
> > > * Sean Busbey ([hidden email])
> > >
> > > == Nominated Mentors ==
> > >
> > > * Bikas Saha ([hidden email])
> > > * Brock Noland ([hidden email])
> >
> >
> > A couple of points:
> >
> > - Sean, while the champion and mentor roles are separate, we do hope that
> > all champions will continue on as a mentor. If this is your intention
> > please add yourself.
>
> After having to withdraw from mentoring a couple of podlings at the
> end of last year I am conservative about what my volunteer time looks
> like right now. I'm certain I can spare the time to help the Livy
> community get introduced to the incubator. I'm not certain beyond
> that, so I am not listed as a formal mentor.
>
> >  - All mentors must be on the IPMC. Foundation membership isn't a
> > requirement, however most people use membership to get access to the
> IPMC.
> > If Bikas wants to be a mentor, he'll need to join the IPMC otherwise
> you'll
> > need to find 2 mentors.
>
> I was pretty sure Bikas had already done this step. I'll chase this
> down and find the disconnect.
>
> > - Do the proposed mentors have a relationship to the incoming project,
> e.g.
> > do they care if it succeeds from a corporate interest standpoint?
>
> I'll let the mentors answer for themselves here, as I won't presume to
> know their specific motivations for volunteering.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>
>
Reply | Threaded
Open this post in threaded view
|

Re: [PROPOSAL] Livy Proposal for Apache Incubator

Sean Busbey-2
Hi folks!

Thanks for all the feedback and on the additional mentor; I think the proposal is ready to go to a vote now.

Presuming folks don't have any last minute questions, I'll post it later today.

-busbey

On 2017-05-23 19:52 (-0500), Kostas Sakellis <[hidden email]> wrote:

> Thank you John for the feedback.
>
> On Mon, May 22, 2017 at 12:30 PM, Sean Busbey <[hidden email]> wrote:
>
> > On 2017-05-21 09:46 (-0500), "John D. Ament" <[hidden email]>
> > wrote:
> > >
> > > On Fri, May 19, 2017 at 7:45 PM Sean Busbey <[hidden email]> wrote:
> > > >
> > > > == Reliance on salaried Developers ==
> > > >
> > > > The existing contributors to the Livy project have been made by
> > salaried
> > > > engineers from Cloudera, Microsoft and Hortonworks. Since there are
> > three
> > > > major organisations involved, the risk of reliance on a single group of
> > > > salaried developers is mitigated. The Livy user base is diverse, with
> > users
> > > > from across the globe, including users from academic settings. We aim
> > to
> > > > further diversify the Livy user and contributor base.
> > > >
> > >
> > > There's a disconnect between this paragraph and the initial committers
> > > list. Specifically, no one from Microsoft is represented (as best as I
> > can
> > > tell).
> >
> > Ah, this lack of clarity is my fault as an editor. One of the initial
> > committers was recently employed by Microsoft but is now in the
> > process of changing employers. Another person formally affiliated with
> > both the project and Microsoft has decided not to continue
> > participating.
> >
> > I could rephrase this to talk about the contributions made to date as
> > being from individuals then in the employ of three major companies.
> > Then call out the initial committer list as from two of those and one
> > unaffiliated. Would that read clearer?
> >
>
> I modified this section in the proposal to read: "The contributions to the
> Livy project to date have been made by salaried engineers from Cloudera,
> Microsoft and Hortonworks. One of the individuals on the initial committer
> list has since left Microsoft and is currently unaffiliated. The remaining
> contributors are from Cloudera and Hortonworks. Since there are at least
> two major organizations involved, the risk of reliance on a single group of
> salaried developers is mitigated. The Livy user base is diverse, with users
> from across the globe, including users from academic settings. We aim to
> further diversify the Livy user and contributor base."
>
>
> >
> > > > Cloudera currently owns the domain name: http://livy.io/ which will be
> > > > transferred to the ASF and redirected to the official page during
> > > > incubation.
> > > >
> > > >
> > >
> > > I'm assuming that the incoming project is aware that we expect the main
> > dev
> > > landing page to be livy.incubator.apache.org . We will want to track
> > this
> > > as a project specific item.
> >
> >
> > Yep, once all the docs are moved over to ASF infrastructure we can
> > just have the current domain act as a redirect. Should I call this out
> > in the proposal?
> >
> > I modified the proposal to say: "Cloudera currently owns the domain name:
> http://livy.io/. Once all the documentation has moved over to ASF
> infrastructure, the main landing page will become livy.incubator.apache.org
> and the old domain will just act as a redirect."
>
>
> >
> > > > == Git Repository ==
> > > >
> > > > git://git.apache.org/livy
> > > >
> > >
> > > Just to confirm - it'll be incubator-livy, not just livy.
> >
> > right right. I'll correct this when making the other edits.
> >
> > Added the correction to the proposal
>
>
> >
> > > > == Issue Tracking ==
> > > >
> > > > We would like to import our current JIRA project into the ASF JIRA,
> > such
> > > > that our historical commit message and code comments continue to
> > reference
> > > > the appropriate bug numbers.
> > > >
> > >
> > > I would recommend reaching out to infra to see if the import is possible
> > > before voting on the project. Otherwise you'll need to list out an la
> >
> > Sure I can reach out. I've seen this done a few times, so I consider
> > it low risk. The end of your line appears to have been lost, what's
> > the "Otherwise..." ?
> >
> > > > = Sponsors =
> > > > == Champion ==
> > > >
> > > > * Sean Busbey ([hidden email])
> > > >
> > > > == Nominated Mentors ==
> > > >
> > > > * Bikas Saha ([hidden email])
> > > > * Brock Noland ([hidden email])
> > >
> > >
> > > A couple of points:
> > >
> > > - Sean, while the champion and mentor roles are separate, we do hope that
> > > all champions will continue on as a mentor. If this is your intention
> > > please add yourself.
> >
> > After having to withdraw from mentoring a couple of podlings at the
> > end of last year I am conservative about what my volunteer time looks
> > like right now. I'm certain I can spare the time to help the Livy
> > community get introduced to the incubator. I'm not certain beyond
> > that, so I am not listed as a formal mentor.
> >
> > >  - All mentors must be on the IPMC. Foundation membership isn't a
> > > requirement, however most people use membership to get access to the
> > IPMC.
> > > If Bikas wants to be a mentor, he'll need to join the IPMC otherwise
> > you'll
> > > need to find 2 mentors.
> >
> > I was pretty sure Bikas had already done this step. I'll chase this
> > down and find the disconnect.
> >
> > > - Do the proposed mentors have a relationship to the incoming project,
> > e.g.
> > > do they care if it succeeds from a corporate interest standpoint?
> >
> > I'll let the mentors answer for themselves here, as I won't presume to
> > know their specific motivations for volunteering.
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [hidden email]
> > For additional commands, e-mail: [hidden email]
> >
> >
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]