[
https://issues.apache.org/jira/browse/FLINK-15171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16992648#comment-16992648
]
Piotr Nowojski edited comment on FLINK-15171 at 12/10/19 3:23 PM:
------------------------------------------------------------------
{noformat}
model name : Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz
{noformat}
{noformat}
MemTotal: 65765564 kB
MemFree: 11099068 kB
MemAvailable: 60741468 kB
Buffers: 2633220 kB
Cached: 36536872 kB
SwapCached: 4 kB
Active: 19142348 kB
Inactive: 20350104 kB
Active(anon): 653616 kB
Inactive(anon): 355320 kB
Active(file): 18488732 kB
Inactive(file): 19994784 kB
Unevictable: 0 kB
Mlocked: 0 kB
SwapTotal: 33521660 kB
SwapFree: 33521616 kB
Dirty: 44 kB
Writeback: 0 kB
AnonPages: 316104 kB
Mapped: 67060 kB
Shmem: 686580 kB
Slab: 15012660 kB
SReclaimable: 11884900 kB
SUnreclaim: 3127760 kB
KernelStack: 3376 kB
PageTables: 4192 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
WritebackTmp: 0 kB
CommitLimit: 66404440 kB
Committed_AS: 2185680 kB
VmallocTotal: 34359738367 kB
VmallocUsed: 0 kB
VmallocChunk: 0 kB
HardwareCorrupted: 0 kB
AnonHugePages: 247808 kB
ShmemHugePages: 0 kB
ShmemPmdMapped: 0 kB
CmaTotal: 0 kB
CmaFree: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
DirectMap4k: 145096 kB
DirectMap2M: 29003776 kB
DirectMap1G: 37748736 kB
{noformat}
https://pastebin.com/9TmcDQQG
I wouldn't rule out some benchmarking issue like the one I mentioned:
{quote}
(changing parallelism in one benchmarks is carried on to the next one?)
{quote}
(as your benchmarking PR changed parallelism of one of the benchmarks). I will
try to take a look and help investigate this tomorrow.
was (Author: pnowojski):
{noformat}
model name : Intel(R) Core(TM) i7-7700 CPU @ 3.60GHz
{noformat}
{noformat}
MemTotal: 65765564 kB
MemFree: 11099068 kB
MemAvailable: 60741468 kB
Buffers: 2633220 kB
Cached: 36536872 kB
SwapCached: 4 kB
Active: 19142348 kB
Inactive: 20350104 kB
Active(anon): 653616 kB
Inactive(anon): 355320 kB
Active(file): 18488732 kB
Inactive(file): 19994784 kB
Unevictable: 0 kB
Mlocked: 0 kB
SwapTotal: 33521660 kB
SwapFree: 33521616 kB
Dirty: 44 kB
Writeback: 0 kB
AnonPages: 316104 kB
Mapped: 67060 kB
Shmem: 686580 kB
Slab: 15012660 kB
SReclaimable: 11884900 kB
SUnreclaim: 3127760 kB
KernelStack: 3376 kB
PageTables: 4192 kB
NFS_Unstable: 0 kB
Bounce: 0 kB
WritebackTmp: 0 kB
CommitLimit: 66404440 kB
Committed_AS: 2185680 kB
VmallocTotal: 34359738367 kB
VmallocUsed: 0 kB
VmallocChunk: 0 kB
HardwareCorrupted: 0 kB
AnonHugePages: 247808 kB
ShmemHugePages: 0 kB
ShmemPmdMapped: 0 kB
CmaTotal: 0 kB
CmaFree: 0 kB
HugePages_Total: 0
HugePages_Free: 0
HugePages_Rsvd: 0
HugePages_Surp: 0
Hugepagesize: 2048 kB
DirectMap4k: 145096 kB
DirectMap2M: 29003776 kB
DirectMap1G: 37748736 kB
{noformat}
https://pastebin.com/9TmcDQQG
I wouldn't rule out some benchmarking issue like the one I mentioned:
{quote}
(changing parallelism in one benchmarks is carried on to the next one?)
{quote}
I will try to take a look and help investigate this tomorrow.
> Performance regression in serialisation benchmarks
> --------------------------------------------------
>
> Key: FLINK-15171
> URL: https://issues.apache.org/jira/browse/FLINK-15171
> Project: Flink
> Issue Type: Bug
> Components: API / Type Serialization System, Benchmarks
> Affects Versions: 1.10.0
> Reporter: Piotr Nowojski
> Priority: Blocker
> Fix For: 1.10.0
>
>
> There is quite significant performance regression in serialisation benchmarks
> in the commit range 2ecf7ca..9320f34 (which includes FLINK-14346).
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerTuple&env=2
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerRow&env=2
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerPojo&env=2
> it coincides with the performance improvement for heavy strings
> http://codespeed.dak8s.net:8000/timeline/?ben=serializerHeavyString&env=2
> it might be caused by some accidental change in the benchmarking code
> (changing parallelism in one benchmarks is carried on to the next one?) or in
> the code itself.
> CC [~rgrebennikov] [~AHeise]
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
