-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/69097/
-----------------------------------------------------------
(Updated Dec. 5, 2018, 5:16 p.m.)
Review request for mesos, Benjamin Mahler and Gastón Kleiman.
Changes
-------
Introduced more chopping in the benchmark and updated the test result.
Bugs: MESOS-6630
https://issues.apache.org/jira/browse/MESOS-6630
Repository: mesos
Description
-------
This benchmark evaluates the allocator performance in
the presence of roles with both small quota (which can
be satisfied by half an agent) as well as large quota
(which need resources from two agents). We setup the cluster,
trigger one allocation cycle and measure the elapsed time.
Diffs (updated)
-----
src/tests/hierarchical_allocator_benchmarks.cpp
6c6330e8cdbc705be322a7e2445b295c35ee6952
Diff: https://reviews.apache.org/r/69097/diff/2/
Changes: https://reviews.apache.org/r/69097/diff/1-2/
Testing (updated)
-------
Result from optimized build on a multicore 2.2GHz machine:
Added 30 agents in 1.612314ms
Added 30 frameworks in 8.353056ms
Benchmark setup: 30 agents, 30 roles, 30 frameworks
Start allocation
Made 36 allocations in 13.844557ms
Made 0 allocation in 8.541409ms
Added 300 agents in 12.506766ms
Added 300 frameworks in 540.262526ms
Benchmark setup: 300 agents, 300 roles, 300 frameworks
Start allocation
Made 350 allocations in 952.103434ms
Made 0 allocation in 770.261096ms
Added 3000 agents in 119.947513ms
Added 3000 frameworks in 55.630070532secs
Benchmark setup: 3000 agents, 3000 roles, 3000 frameworks
Start allocation
Made 3500 allocations in 1.974348067mins
Made 0 allocation in 1.83595374028333mins
Thanks,
Meng Zhu
