[
https://issues.apache.org/jira/browse/FLINK-20612?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17250863#comment-17250863
]
Zhu Zhu commented on FLINK-20612:
---------------------------------
Thanks for proposing this [~Thesharing] and sharing the statistics. Benchmarks
for scheduler sounds a really good idea.
It can be helpful to detect performance issues, like the long time for
scheduling strategy initialization which is unexpected to me, and the
deployment slowness which was also reported in FLINK-16069 .
I think targeting it first for `flink-benchmarks` is better to identify
performance regressions. For Unit test we may have to set a much larger timeout
to avoid unstable tests, and regressions can be covered as a result.
However, I think it's still good to have some CI perf tests for critical
process when improving its performance, so that big performance issues can be
found before change merged and it can also allows to quick check the current
performance status in development process. Maybe we can have some common utils
in Flink scheduler tests, so that both UT and flink-benchmarks can make use of
them and we can avoid developing the same things twice?
> Add benchmarks for scheduler
> ----------------------------
>
> Key: FLINK-20612
> URL: https://issues.apache.org/jira/browse/FLINK-20612
> Project: Flink
> Issue Type: Improvement
> Components: Runtime / Coordination
> Affects Versions: 1.13.0
> Reporter: Zhilong Hong
> Priority: Major
>
> With Flink 1.12, we failed to run large-scale jobs on our cluster. When we
> were trying to run the jobs, we met the exceptions like out of heap memory,
> taskmanager heartbeat timeout, and etc. We increased the size of heap memory
> and extended the heartbeat timeout, the job still failed. After the
> troubleshooting, we found that there are some performance bottlenecks in the
> jobmaster. These bottlenecks are highly related to the complexity of the
> topology.
> We implemented several benchmarks on these bottlenecks based on
> flink-benchmark. The topology of the benchmarks is a simple graph, which
> consists of only two vertices: one source vertex and one sink vertex. They
> are both connected with all-to-all blocking edges. The parallelisms of the
> vertices are both 8000. The execution mode is batch. The results of the
> benchmarks are illustrated below:
> Table 1: The result of benchmarks on bottlenecks in the jobmaster
> | |*Time spent*|
> |Build topology|19970.44 ms|
> |Init scheduling strategy|38167.351 ms|
> |Deploy tasks|15102.850 ms|
> |Calculate failover region to restart|12080.271 ms|
> We'd like to propose these benchmarks for procedures related to the
> scheduler. There are three main benefits:
> # They help us to understand the current status of task deployment
> performance and locate where the bottleneck is.
> # We can use the benchmarks to evaluate the optimization in the future.
> # As we run the benchmarks daily, they will help us to trace how the
> performance changes and locate the commit that introduces the performance
> regression if there is any.
> In the first version of the benchmarks, we mainly focus on the procedures we
> mentioned above. The methods corresponding to the procedures are:
> # Building topology: {{ExecutionGraph#attachJobGraph}}
> # Initializing scheduling strategies:
> {{PipelinedRegionSchedulingStrategy#init}}
> # Deploying tasks: {{Execution#deploy}}
> # Calculating failover regions:
> {{RestartPipelinedRegionFailoverStrategy#getTasksNeedingRestart}}
> In the benchmarks, the topology consists of two vertices: source -> sink.
> They are connected with all-to-all edges. The result partition type
> ({{PIPELINED}} and {{BLOCKING}}) should be considered separately.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
