mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jiang Yan Xu <...@jxu.me>
Subject Re: Review Request 49616: Add suppression benchmark.
Date Thu, 28 Jul 2016 16:02:26 GMT

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/49616/#review143901
-----------------------------------------------------------


Fix it, then Ship it!





src/tests/hierarchical_allocator_tests.cpp (lines 3808 - 3809)
<https://reviews.apache.org/r/49616/#comment209861>

    Move this to right above `agentTotal`.



src/tests/hierarchical_allocator_tests.cpp (line 3813)
<https://reviews.apache.org/r/49616/#comment209859>

    The asymmetry between the string typed `agentTotal` and the `Resources` typed `allocation`
looks jarring to me, how about just:
    
    ```
    SlaveInfo agent = createSlaveInfo("cpus:24;mem:4096;disk:4096;ports:[31000-32000]");
    
    // Then later,
    
    agents.push_back(agent);
    ```



src/tests/hierarchical_allocator_tests.cpp (line 3847)
<https://reviews.apache.org/r/49616/#comment209891>

    The TODO would be to parameterize the `allocationsCount` right?



src/tests/hierarchical_allocator_tests.cpp (line 3855)
<https://reviews.apache.org/r/49616/#comment209862>

    const Allocation& allocation



src/tests/hierarchical_allocator_tests.cpp (lines 3868 - 3869)
<https://reviews.apache.org/r/49616/#comment209893>

    The implication is that there can be `frameworkCount % allocationsCount` frameworks not
suppressed, which is fine in our test. Let's add a comment on this.
    
    Note that we could always choose `allocationsCount`s so `frameworkCount % allocationsCount
== 0` even when we parameterize it but still we are fine even if it's not the case.



src/tests/hierarchical_allocator_tests.cpp (line 3874)
<https://reviews.apache.org/r/49616/#comment209892>

    We need a settle right before `watch.start()` so we are truely only measuring the allocation
time.


- Jiang Yan Xu


On July 26, 2016, 12:28 a.m., Jacob Janco wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/49616/
> -----------------------------------------------------------
> 
> (Updated July 26, 2016, 12:28 a.m.)
> 
> 
> Review request for mesos, James Peach, Joris Van Remoortere, and Jiang Yan Xu.
> 
> 
> Bugs: MESOS-5781
>     https://issues.apache.org/jira/browse/MESOS-5781
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> - Useful for high framework clusters which carry
>   many suppressed frameworks that are considered
>   during allocation.
> 
> 
> Diffs
> -----
> 
>   src/tests/hierarchical_allocator_tests.cpp bb6947fcecb5b78047e98d10fc1278c612a69548

> 
> Diff: https://reviews.apache.org/r/49616/diff/
> 
> 
> Testing
> -------
> 
> MESOS_BENCHMARK=1 GTEST_FILTER="*BENCHMARK_Test.SuppressOffers*" make check
> 
> 
> Thanks,
> 
> Jacob Janco
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message