-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/43686/
-----------------------------------------------------------

Review request for mesos, Joris Van Remoortere and Michael Park.


Bugs: MESOS-4691
    https://issues.apache.org/jira/browse/MESOS-4691


Repository: mesos


Description
-------

For the particular workload exercised by the benchmark, this suggests that
adding a 12-element label to a resource slows down allocation by about 5% on my
local machine.


Diffs
-----

  src/tests/hierarchical_allocator_tests.cpp 
990f3723d52dfeaa19d5eb0603c0fc7eb2b362c7 

Diff: https://reviews.apache.org/r/43686/diff/


Testing
-------

make check

FYI, results on my laptop:

_Original benchmark (unlabeled resources)_
[ RUN      ] HierarchicalAllocator_BENCHMARK_Test.DeclineOffers
Using 2000 slaves and 200 frameworks
round 0 allocate took 2.028175secs to make 200 offers
round 1 allocate took 2.006791secs to make 200 offers
round 2 allocate took 2.033723secs to make 200 offers
round 3 allocate took 2.017508secs to make 200 offers
round 4 allocate took 2.037235secs to make 200 offers
round 5 allocate took 2.054095secs to make 200 offers
round 6 allocate took 2.048884secs to make 200 offers
round 7 allocate took 2.044252secs to make 200 offers
round 8 allocate took 2.060256secs to make 200 offers
round 9 allocate took 2.07121secs to make 200 offers
round 10 allocate took 2.066261secs to make 200 offers
round 11 allocate took 2.034805secs to make 200 offers
round 12 allocate took 2.053705secs to make 200 offers
round 13 allocate took 2.042106secs to make 200 offers
round 14 allocate took 2.082704secs to make 200 offers

_New benchmark (two labeled resources with different labels)_
[ RUN      ] HierarchicalAllocator_BENCHMARK_Test.ResourceLabels
Using 2000 slaves and 200 frameworks
round 0 allocate took 2.128709secs to make 200 offers
round 1 allocate took 2.188029secs to make 200 offers
round 2 allocate took 2.145937secs to make 200 offers
round 3 allocate took 2.171442secs to make 200 offers
round 4 allocate took 2.153106secs to make 200 offers
round 5 allocate took 2.151484secs to make 200 offers
round 6 allocate took 2.136182secs to make 200 offers
round 7 allocate took 2.152105secs to make 200 offers
round 8 allocate took 2.187842secs to make 200 offers
round 9 allocate took 2.13839secs to make 200 offers
round 10 allocate took 2.237216secs to make 200 offers
round 11 allocate took 2.164702secs to make 200 offers
round 12 allocate took 2.143296secs to make 200 offers
round 13 allocate took 2.198839secs to make 200 offers
round 14 allocate took 2.179931secs to make 200 offers

For fun, I tried running the benchmark with the modified equality operator for 
`Labels` that uses `unordered_multiset` to produce the correct results for 
labels that contain duplicates:

[ RUN      ] HierarchicalAllocator_BENCHMARK_Test.ResourceLabels
Using 2000 slaves and 200 frameworks
round 0 allocate took 2.190051secs to make 200 offers
round 1 allocate took 2.169332secs to make 200 offers
round 2 allocate took 2.156235secs to make 200 offers
round 3 allocate took 2.15506secs to make 200 offers
round 4 allocate took 2.133953secs to make 200 offers
round 5 allocate took 2.18325secs to make 200 offers
round 6 allocate took 2.164478secs to make 200 offers
round 7 allocate took 2.192077secs to make 200 offers
round 8 allocate took 2.14688secs to make 200 offers
round 9 allocate took 2.172333secs to make 200 offers
round 10 allocate took 2.199906secs to make 200 offers
round 11 allocate took 2.16384secs to make 200 offers
round 12 allocate took 2.200181secs to make 200 offers
round 13 allocate took 2.138463secs to make 200 offers
round 14 allocate took 2.184699secs to make 200 offers


Thanks,

Neil Conway

Reply via email to