[jira] [Commented] (SPARK-16367) Wheelhouse Support for PySpark

["Jeff Zhang (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/SPARK-16367?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15362764#comment-15362764 ] 

Jeff Zhang commented on SPARK-16367:
------------------------------------

[~gaetan@xeberon.net] Thanks for the new idea, this makes the community more powerful. 
Here's my few concerns and comments.

bq. support of heterogenous Spark nodes (ex: 32 bits, 64 bits) is possible but one has to send all wheels flavours and ensure pip is able to install in every environment.
What about different OS ? Can Wheels compiled on the client machine be used on a different OS ? This is my largest concern for this approach. 

bq. and disk space
For yarn, this is not a problem, because the container will be cleanup after the container is exited.

I am not sure whether the extra steps for creating wheelhouse is too complicated for users. 
BTW in the approach of SPARK-13587, I specified the cache-dir when creating virtualenv. That means only the first time compilation is needed, after that each installation will pick up the wheel file in cache dir. 


> Wheelhouse Support for PySpark
> ------------------------------
>
>                 Key: SPARK-16367
>                 URL: https://issues.apache.org/jira/browse/SPARK-16367
>             Project: Spark
>          Issue Type: New Feature
>          Components: Deploy, PySpark
>    Affects Versions: 1.6.1, 1.6.2, 2.0.0
>            Reporter: Semet
>              Labels: newbie, python, python-wheel, wheelhouse
>   Original Estimate: 168h
>  Remaining Estimate: 168h
>
> *Rational*
> Is it recommended, in order to deploying Scala packages written in Scala, to build big fat jar files. This allows to have all dependencies on one package so the only "cost" is copy time to deploy this file on every Spark Node.
> On the other hand, Python deployment is more difficult once you want to use external packages, and you don't really want to mess with the IT to deploy the packages on the virtualenv of each nodes.
> *Previous approaches*
> I based the current proposal over the two following bugs related to this point:
> - SPARK-6764 ("Wheel support for PySpark")
> - SPARK-13587("Support virtualenv in PySpark")
> First part of my proposal was to merge, in order to support wheels install and virtualenv creation
> *Uber Fat Wheelhouse for Python Deployment*
> In Python, the packaging standard is now "wheels", which goes further that old good ".egg" files. With a wheel file (".whl"), the package is already prepared for a given architecture. You can have several wheel, each specific to an architecture, or environment. 
> The {{pip}} tools now how to select the package matching the current system, how to install this package in a light speed. Said otherwise, package that requires compilation of a C module, for instance, does *not* compile anything when installing from wheel file.
> {{pip}} also provides the ability to generate easily all wheel of all packages used for a given module (inside a "virtualenv"). This is called "wheelhouse". You can even don't mess with this compilation and retrieve it directly from pypi.python.org.
> *Developer workflow*
> Here is, in a more concrete way, my proposal for on Pyspark developers point of view:
> - you are writing a PySpark script that increase in term of size and dependencies. Deploying on Spark for example requires to build numpy or Theano and other dependencies
> - to use "Big Fat Wheelhouse" support of Pyspark, you need to turn his script into a standard Python package:
> -- write a {{requirements.txt}}. I recommend to specify all package version. You can use [pip-tools|https://github.com/nvie/pip-tools] to maintain the requirements.txt
> {code}
> astroid==1.4.6            # via pylint
> autopep8==1.2.4
> click==6.6                # via pip-tools
> colorama==0.3.7           # via pylint
> enum34==1.1.6             # via hypothesis
> findspark==1.0.0          # via spark-testing-base
> first==2.0.1              # via pip-tools
> hypothesis==3.4.0         # via spark-testing-base
> lazy-object-proxy==1.2.2  # via astroid
> linecache2==1.0.0         # via traceback2
> pbr==1.10.0
> pep8==1.7.0               # via autopep8
> pip-tools==1.6.5
> py==1.4.31                # via pytest
> pyflakes==1.2.3
> pylint==1.5.6
> pytest==2.9.2             # via spark-testing-base
> six==1.10.0               # via astroid, pip-tools, pylint, unittest2
> spark-testing-base==0.0.7.post2
> traceback2==1.4.0         # via unittest2
> unittest2==1.1.0          # via spark-testing-base
> wheel==0.29.0
> wrapt==1.10.8             # via astroid
> {code}
> -- write a setup.py with some entry points or package. Use [PBR|http://docs.openstack.org/developer/pbr/] it makes the jobs of maitaining a setup.py files really easy
> -- create a virtualenv if not already in one:
> {code}
> virtualenv env
> {code}
> -- Work on your environment, define the requirement you need in {{requirements.txt}}, do all the {{pip install}} you need.
> - create the wheelhouse for your current project
> {code}
> pip install wheelhouse
> pip wheel . --wheel-dir wheelhouse
> {code}
> This can take some times, but at the end you have all the .whl required *for your current system*
> - zip it into a {{wheelhouse.zip}}.
> Note that you can have your own package (for instance 'my_package') be generated into a wheel and so installed by {{pip}} automatically.
> Now comes the time to submit the project:
> {code}
> bin/spark-submit  --master master --deploy-mode client --files /path/to/virtualenv/requirements.txt,/path/to/virtualenv/wheelhouse.zip --conf "spark.pyspark.virtualenv.enabled=true" ~/path/to/launcher_script.py
> {code}
> You can see that:
> - no extra argument is add in the command line. All configuration goes through {{--conf}} argument (this has been directly taken from SPARK-13587). According to the history on spark source code, I guess the goal is to simplify the maintainance of the various command line interface, by avoiding too many specific argument.
> - The wheelhouse deployment is triggered by the {{ --conf "spark.pyspark.virtualenv.enabled=true" }} argument. The {{requirements.txt}} and {{wheelhouse.zip}} are copied through {{--files}}. The names of both files can be changed through {{--conf}} arguments. I guess with a proper documentation this might not be a problem
> - you still need to define the path to {{requirement.txt}} and {{wheelhouse.zip}} (they will be automatically copied to each node). This is important since this will allow {{pip install}}, running of each node, to pick only the wheels he needs. For example, if you have a package compiled on 32 bits and 64 bits, you will have 2 wheels, and on each node, {{pip}} will only select the right one
> - I have choosen to keep the script at the end of the command line, but for me it is just a launcher script, it can only be 4 lines:
> {code}
> /#!/usr/bin/env python	
> from mypackage import run
> run()
> {code}
> - on each node, a new virtualenv is created *at each deployment*. This has a cost, but not so much, since the {{pip install}} will only install wheel, no compilation nor internet connection will be required. The command line for installing the wheel on each node will be like: 
> {code}
> pip install --no-index --find-links=/path/to/node/wheelhouse -r requirements.txt
> {code}
> *advantages*
> - quick installation, since there is no compilation
> - no Internet connectivity support, no need mess with the corporate proxy or require a local mirroring of pypi.
> - package versionning isolation (two spark job can depends on two different version of a given library)
> *disadvantages*
> - creating a virtualenv at each execution takes time, not that much but still it can take some seconds
> - and disk space
> - slighly more complex to setup than sending a simple python script, but this feature is not lost
> - support of heterogenous Spark nodes (ex: 32 bits, 64 bits) is possible but one has to send all wheels flavours and ensure pip is able to install in every environment. The complexity of this task is on the hands of the developer and no more on the IT persons! (TMHO, this is an advantage)
> *code submission*
> I already started working on this point, starting by merging the 2 mergerequests [#5408|https://github.com/apache/spark/pull/5408] and [#13599|https://github.com/apache/spark/pull/13599]
> I'll upload a patch asap for review.
> I see two major interogations:
> - I don't know that much YARN or MESOS, so I might require some help for the final integration
> - documentation should really be carefully crafted so users are not lost in all these concepts
> I really think having this "wheelhouse" support for spark will really helps using, maintaining, and evolving Python scripts on Spark. Python has a rich set of mature libraries Spark should do anythink to help developers easily access and use them in their everyday job.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org