[
https://issues.apache.org/jira/browse/FLINK-838?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14065572#comment-14065572
]
Fabian Hueske commented on FLINK-838:
-------------------------------------
Controlling the number of Map tasks in Hadoop is a bit tricky.
In fact, the number of map tasks is given to the InputFormat when creating the
InputSplits. For DFS files, the number of blocks is the lower bound of input
splits though. A block can be split into multiple splits but it is not possible
to read multiple blocks in one split. The rational is to reduce remote reading
of blocks.
Therefore, you have always at least as many Map tasks as your input has DFS
blocks. It is possible to split the input into smaller splits though.
http://wiki.apache.org/hadoop/HowManyMapsAndReduces gives a short overview.
I suggested to have an upper bound for Map and Reduce tasks in Flink to control
the number of concurrently running tasks. For Map this parameter would have a
similar effect as the number of Map slots. For Reduce the parameter would have
a somewhat different effect, because in Hadoop you can reduce the size of
Reduce tasks by increasing their number which does not work for Flink if you
limit the number of concurrently running tasks.
> GSoC Summer Project: Implement full Hadoop Compatibility Layer for
> Stratosphere
> -------------------------------------------------------------------------------
>
> Key: FLINK-838
> URL: https://issues.apache.org/jira/browse/FLINK-838
> Project: Flink
> Issue Type: Improvement
> Reporter: GitHub Import
> Assignee: Fabian Hueske
> Labels: github-import
> Fix For: pre-apache
>
>
> This is a meta issue for tracking @atsikiridis progress with implementing a
> full Hadoop Compatibliltiy Layer for Stratosphere.
> Some documentation can be found in the Wiki:
> https://github.com/stratosphere/stratosphere/wiki/%5BGSoC-14%5D-A-Hadoop-abstraction-layer-for-Stratosphere-(Project-Map-and-Notes)
> As well as the project proposal:
> https://github.com/stratosphere/stratosphere/wiki/GSoC-2014-Project-Proposal-Draft-by-Artem-Tsikiridis
> Most importantly, there is the following **schedule**:
> *19 May - 27 June (Midterm)*
> 1) Work on the Hadoop tasks, their Context and the mapping of Hadoop's
> Configuration to the one of Stratosphere. By successfully bridging the Hadoop
> tasks with Stratosphere, we already cover the most basic Hadoop Jobs. This
> can be determined by running some popular Hadoop examples on Stratosphere
> (e.g. WordCount, k-means, join) (4 - 5 weeks)
> 2) Understand how the running of these jobs works (e.g. command line
> interface) for the wrapper. Implement how will the user run them. (1 - 2
> weeks).
> *27 June - 11 August*
> 1) Continue wrapping more "advanced" Hadoop Interfaces (Comparators,
> Partitioners, Distributed Cache etc.) There are quite a few interfaces and it
> will be a challenge to support all of them. (5 full weeks)
> 2) Profiling of the application and optimizations (if applicable)
> *11 August - 18 August*
> Write documentation on code, write a README with care and add more
> unit-tests. (1 week)
> ---------------- Imported from GitHub ----------------
> Url: https://github.com/stratosphere/stratosphere/issues/838
> Created by: [rmetzger|https://github.com/rmetzger]
> Labels: core, enhancement, parent-for-major-feature,
> Milestone: Release 0.7 (unplanned)
> Created at: Tue May 20 10:11:34 CEST 2014
> State: open
--
This message was sent by Atlassian JIRA
(v6.2#6252)
