mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Anand Mazumdar <an...@apache.org>
Subject Re: Review Request 53113: Added benchmark for measuring call ingestion throughput.
Date Thu, 27 Oct 2016 02:45:18 GMT


> On Oct. 25, 2016, 8:28 p.m., Vinod Kone wrote:
> > Can you paste numbers for the benchmark in the testing section for posterity?

My bad, fixed.


> On Oct. 25, 2016, 8:28 p.m., Vinod Kone wrote:
> > src/tests/scheduler_tests.cpp, line 1822
> > <https://reviews.apache.org/r/53113/diff/1/?file=1543713#file1543713line1822>
> >
> >     why do you say driver here?

The comment was around what these benchmark tests do i.e., they measure the time to ingest
calls using the scheduler library and driver and was not limited to that particular test.


- Anand


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/53113/#review153820
-----------------------------------------------------------


On Oct. 22, 2016, 10:25 p.m., Anand Mazumdar wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/53113/
> -----------------------------------------------------------
> 
> (Updated Oct. 22, 2016, 10:25 p.m.)
> 
> 
> Review request for mesos and Vinod Kone.
> 
> 
> Bugs: MESOS-6405
>     https://issues.apache.org/jira/browse/MESOS-6405
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> This change adds a benchmark for measuring the time taken by
> the master to process calls using the scheduler library/driver.
> The performance fixes for any regressions seen would be made in
> follow up patches.
> 
> 
> Diffs
> -----
> 
>   src/tests/scheduler_tests.cpp abd1f61823a683cefe8c66abf176eba7d9d26b40 
> 
> Diff: https://reviews.apache.org/r/53113/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> 
> Thanks,
> 
> Anand Mazumdar
> 
>


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