> On June 19, 2015, 10:31 p.m., Ben Mahler wrote:
> > src/master/allocator/sorter/drf/sorter.cpp, line 299
> > <https://reviews.apache.org/r/35664/diff/2/?file=988541#file988541line299>
> >
> >     s/scalar/name/ ?

It'll hide the function parameter 'name'. I'll do a sweep in a follow up patch 
to s/name/client/


> On June 19, 2015, 10:31 p.m., Ben Mahler wrote:
> > src/master/allocator/sorter/drf/sorter.cpp, line 317
> > <https://reviews.apache.org/r/35664/diff/2/?file=988541#file988541line317>
> >
> >     Do you want to use 0.0 here and elsewhere to make it clear that these 
> > are doubles?

Done.


> On June 19, 2015, 10:31 p.m., Ben Mahler wrote:
> > src/tests/sorter_tests.cpp, lines 396-398
> > <https://reviews.apache.org/r/35664/diff/2/?file=988542#file988542line396>
> >
> >     Hm.. wonder if we can do the following now in these tests?
> >     
> >     ```
> >     EXPECT_EQ({"a", "b"}, sorter.sort());
> >     ```

Tried, doesn't work:)


- Jie


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/35664/#review88591
-----------------------------------------------------------


On June 19, 2015, 10:21 p.m., Jie Yu wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/35664/
> -----------------------------------------------------------
> 
> (Updated June 19, 2015, 10:21 p.m.)
> 
> 
> Review request for mesos, Ben Mahler and Vinod Kone.
> 
> 
> Bugs: MESOS-2891
>     https://issues.apache.org/jira/browse/MESOS-2891
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> Improved the performance of DRF sorter by caching the scalars.
> 
> 
> Diffs
> -----
> 
>   src/master/allocator/sorter/drf/sorter.hpp 
> 35dc1a4d0b5e61b26a07c2c53583d75896aff27c 
>   src/master/allocator/sorter/drf/sorter.cpp 
> c5f4caf68aff26a9d4809eff11df98d0137aa401 
>   src/tests/sorter_tests.cpp 435e0bfeb28c1a9ea124312a8b97a869945ac87f 
> 
> Diff: https://reviews.apache.org/r/35664/diff/
> 
> 
> Testing
> -------
> 
> make check
> 
> With this patch:
> 
> [==========] Running 6 tests from 1 test case.
> [----------] Global test environment set-up.
> [----------] 6 tests from SlaveCount/HierarchicalAllocator_BENCHMARK_Test
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/0
> Added 1000 slaves in 668.950475ms
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/0 (844 
> ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/1
> Added 5000 slaves in 3.335592974secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/1 (4244 
> ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/2
> Added 10000 slaves in 6.675837646secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/2 (8527 
> ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/3
> Added 20000 slaves in 13.411382999secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/3 
> (17177 ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/4
> Added 30000 slaves in 20.012000768secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/4 
> (25567 ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/5
> Added 50000 slaves in 33.377255617secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/5 
> (42960 ms)
> [----------] 6 tests from SlaveCount/HierarchicalAllocator_BENCHMARK_Test 
> (99346 ms total)
> 
> [----------] Global test environment tear-down
> [==========] 6 tests from 1 test case ran. (99452 ms total)
> [  PASSED  ] 6 tests.
> 
> Baseline (pre MESOS-2373)
> 
> [==========] Running 6 tests from 1 test case.
> [----------] Global test environment set-up.
> [----------] 6 tests from SlaveCount/HierarchicalAllocator_BENCHMARK_Test
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/0
> Added 1000 slaves in 492.832376ms
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/0 (602 
> ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/1
> Added 5000 slaves in 2.432903553secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/1 (2991 
> ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/2
> Added 10000 slaves in 4.866961208secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/2 (6014 
> ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/3
> Added 20000 slaves in 9.736516799secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/3 
> (12067 ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/4
> Added 30000 slaves in 14.599861839secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/4 
> (18011 ms)
> [ RUN      ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/5
> Added 50000 slaves in 24.30015991secs
> [       OK ] SlaveCount/HierarchicalAllocator_BENCHMARK_Test.AddSlave/5 
> (30097 ms)
> [----------] 6 tests from SlaveCount/HierarchicalAllocator_BENCHMARK_Test 
> (69784 ms total)
> 
> [----------] Global test environment tear-down
> [==========] 6 tests from 1 test case ran. (69820 ms total)
> [  PASSED  ] 6 tests.
> 
> 
> Thanks,
> 
> Jie Yu
> 
>

Reply via email to