mirror of
https://github.com/netbirdio/gvisor.git
synced 2026-05-22 17:12:49 -07:00
Split fio read/write and randread/randwrite operations
The fio benchmark was changed to a fixed size read/write ammount because the timed benchmark was overwhelming machine memory on tmpfs mounts. Now rand(read|write) operations are prohibitively long, leading to timeouts. Split the benchmarks as they were in python bm-tools: the read/write as fixed sized (1GB) and the rand(read|write) as timed operations (15s). PiperOrigin-RevId: 364584436
This commit is contained in:
committed by
gVisor bot
parent
409a114454
commit
98f378d9ef
@@ -188,8 +188,12 @@ steps:
|
||||
# but for tmpfs mounts, the size can grow to more memory than the machine
|
||||
# has availabe. Fix the runs to 1GB written/read for the benchmark.
|
||||
- <<: *benchmarks
|
||||
label: ":floppy_disk: FIO benchmarks"
|
||||
command: make benchmark-platforms BENCHMARKS_SUITE=fio BENCHMARKS_TARGETS=test/benchmarks/fs:fio_test BENCHMARKS_OPTIONS=--test.benchtime=1000x
|
||||
label: ":floppy_disk: FIO benchmarks (read/write)"
|
||||
command: make benchmark-platforms BENCHMARKS_SUITE=fio BENCHMARKS_TARGETS=test/benchmarks/fs:fio_test BENCHMARKS_FILTER=Fio/operation\.[rw][er] BENCHMARKS_OPTIONS=--test.benchtime=1000x
|
||||
# For rand(read|write) fio benchmarks, running 15s does not overwhelm the system for tmpfs mounts.
|
||||
- <<: *benchmarks
|
||||
label: ":cd: FIO benchmarks (randread/randwrite)"
|
||||
command: make benchmark-platforms BENCHMARKS_SUITE=fio BENCHMARKS_TARGETS=test/benchmarks/fs:fio_test BENCHMARKS_FILTER=Fio/operation\.rand BENCHMARKS_OPTIONS=--test.benchtime=15s
|
||||
- <<: *benchmarks
|
||||
label: ":globe_with_meridians: HTTPD benchmarks"
|
||||
command: make benchmark-platforms BENCHMARKS_FILTER="Continuous" BENCHMARKS_SUITE=httpd BENCHMARKS_TARGETS=test/benchmarks/network:httpd_test
|
||||
|
||||
Reference in New Issue
Block a user