[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20150318050319.GD30894@yliu-dev.sh.intel.com>
Date: Wed, 18 Mar 2015 13:03:19 +0800
From: Yuanahn Liu <yuanhan.liu@...ux.intel.com>
To: "shli@...nel.org" <shli@...nel.org>
Cc: NeilBrown <neilb@...e.de>, lkp@...org, lkp@...ux.intel.com,
linux-kernel@...r.kernel.org, Jaegeuk Kim <jaegeuk@...nel.org>,
linux-f2fs-devel@...ts.sourceforge.net
Subject: performance changes on d4b4c2cd: 37.6% fsmark.files_per_sec, -15.9%
fsmark.files_per_sec, and few more
Hi,
FYI, we noticed performance changes on `fsmark.files_per_sec' by d4b4c2cdffab86f5c7594c44635286a6d277d5c6:
> commit d4b4c2cdffab86f5c7594c44635286a6d277d5c6
> Author: shli@...nel.org <shli@...nel.org>
> AuthorDate: Mon Dec 15 12:57:03 2014 +1100
> Commit: NeilBrown <neilb@...e.de>
> CommitDate: Wed Mar 4 13:40:17 2015 +1100
>
> RAID5: batch adjacent full stripe write
c1dfe87e41d9c2926fe92f803f02c733ddbccf0b d4b4c2cdffab86f5c7594c44635286a6d277d5c6
---------------------------------------- ----------------------------------------
run time(m) metric_value ±stddev run time(m) metric_value ±stddev change testbox/benchmark/sub-testcase
--- ------ ---------------------------- --- ------ ---------------------------- -------- ------------------------------
4 15.3 33.525 ±3.0% 6 11.1 46.133 ±5.0% 37.6% ivb44/fsmark/1x-1t-3HDD-RAID5-xfs-4M-120G-NoSync
3 0.5 262.800 ±1.5% 3 0.4 307.367 ±1.2% 17.0% ivb44/fsmark/1x-1t-4BRD_12G-RAID5-f2fs-4M-30G-NoSync
3 0.5 289.900 ±0.3% 3 0.4 323.367 ±2.4% 11.5% ivb44/fsmark/1x-64t-4BRD_12G-RAID5-f2fs-4M-30G-NoSync
3 0.5 325.667 ±2.2% 3 0.5 358.800 ±1.8% 10.2% ivb44/fsmark/1x-64t-4BRD_12G-RAID5-ext4-4M-30G-NoSync
3 0.6 216.100 ±0.4% 3 0.6 230.100 ±0.4% 6.5% ivb44/fsmark/1x-64t-4BRD_12G-RAID5-f2fs-4M-30G-fsyncBeforeClose
3 0.5 309.900 ±0.3% 3 0.5 328.500 ±1.1% 6.0% ivb44/fsmark/1x-64t-4BRD_12G-RAID5-xfs-4M-30G-NoSync
3 13.8 37.000 ±0.2% 3 16.5 31.100 ±0.3% -15.9% ivb44/fsmark/1x-1t-3HDD-RAID5-f2fs-4M-120G-NoSync
NOTE: here are some more info about those test parameters for you to
understand the testcase better:
1x : where 'x' means iterations or loop, corresponding to the 'L' option of fsmark
64t: where 't' means thread
4M : means the single file size, corresponding to the '-s' option of fsmark
120G, 30G: means the total test size
4BRD_12G: BRD is the ramdisk, where '4' means 4 ramdisk, and where '12G' means
the size of one ramdisk. So, it would be 48G in total. And we made a
raid on those ramdisk.
And FYI, here I listed more detailed changes for the maximal postive and negtive changes.
more detailed changes about ivb44/fsmark/1x-1t-3HDD-RAID5-xfs-4M-120G-NoSync
---------
c1dfe87e41d9c292 d4b4c2cdffab86f5c7594c4463
---------------- --------------------------
%stddev %change %stddev
\ | \
33.53 ± 3% +37.6% 46.13 ± 4% fsmark.files_per_sec
916 ± 3% -27.2% 667 ± 5% fsmark.time.elapsed_time.max
916 ± 3% -27.2% 667 ± 5% fsmark.time.elapsed_time
7 ± 5% +37.6% 10 ± 6% fsmark.time.percent_of_cpu_this_job_got
92097 ± 2% -23.1% 70865 ± 4% fsmark.time.voluntary_context_switches
0.04 ± 42% +681.0% 0.27 ± 22% turbostat.Pkg%pc3
716062 ± 3% -82.7% 124210 ± 21% cpuidle.C1-IVT.usage
6.883e+08 ± 2% -86.8% 91146705 ± 34% cpuidle.C1-IVT.time
0.04 ± 30% +145.8% 0.10 ± 25% turbostat.CPU%c3
404 ± 16% -58.4% 168 ± 14% cpuidle.POLL.usage
159 ± 47% +179.5% 444 ± 23% proc-vmstat.kswapd_low_wmark_hit_quickly
11133 ± 23% +100.3% 22298 ± 30% cpuidle.C3-IVT.usage
10286681 ± 27% +95.6% 20116924 ± 27% cpuidle.C3-IVT.time
7.92 ± 16% +77.4% 14.05 ± 6% turbostat.Pkg%pc6
4.93 ± 3% -38.6% 3.03 ± 2% turbostat.CPU%c1
916 ± 3% -27.2% 667 ± 5% time.elapsed_time.max
916 ± 3% -27.2% 667 ± 5% time.elapsed_time
2137390 ± 3% -26.7% 1566752 ± 5% proc-vmstat.pgfault
7 ± 5% +37.6% 10 ± 6% time.percent_of_cpu_this_job_got
4.309e+10 ± 3% -26.3% 3.176e+10 ± 5% cpuidle.C6-IVT.time
49038 ± 2% -23.9% 37334 ± 4% uptime.idle
1047 ± 2% -23.8% 797 ± 4% uptime.boot
92097 ± 2% -23.1% 70865 ± 4% time.voluntary_context_switches
4005888 ± 0% +13.3% 4537685 ± 11% meminfo.DirectMap2M
3917 ± 2% -16.3% 3278 ± 5% proc-vmstat.pageoutrun
213737 ± 1% -13.9% 183969 ± 3% softirqs.SCHED
46.86 ± 1% +16.5% 54.59 ± 1% turbostat.Pkg%pc2
32603 ± 3% -11.7% 28781 ± 5% numa-vmstat.node1.nr_unevictable
130415 ± 3% -11.7% 115127 ± 5% numa-meminfo.node1.Unevictable
256781 ± 2% -8.8% 234146 ± 3% softirqs.TASKLET
253606 ± 2% -8.9% 231108 ± 3% softirqs.BLOCK
119.10 ± 2% -70.0% 35.78 ± 13% iostat.sdc.rrqm/s
119.86 ± 1% -70.3% 35.64 ± 12% iostat.sdb.rrqm/s
117.13 ± 2% -70.2% 34.96 ± 11% iostat.sda.rrqm/s
504 ± 2% -67.6% 163 ± 12% iostat.sdc.rkB/s
507 ± 1% -67.9% 163 ± 12% iostat.sdb.rkB/s
496 ± 2% -67.7% 160 ± 11% iostat.sda.rkB/s
15392 ± 3% +37.8% 21203 ± 5% iostat.sdb.wrqm/s
15393 ± 3% +37.7% 21203 ± 5% iostat.sdc.wrqm/s
15392 ± 3% +37.7% 21203 ± 5% iostat.sda.wrqm/s
125236 ± 3% +37.7% 172422 ± 4% vmstat.io.bo
125181 ± 3% +37.6% 172303 ± 4% iostat.md0.wkB/s
552 ± 3% +37.6% 760 ± 4% iostat.md0.w/s
62611 ± 3% +37.6% 86167 ± 4% iostat.sdb.wkB/s
62613 ± 3% +37.6% 86167 ± 4% iostat.sdc.wkB/s
62613 ± 3% +37.6% 86168 ± 4% iostat.sda.wkB/s
40.24 ± 1% -18.5% 32.81 ± 2% turbostat.CorWatt
200 ± 0% +22.2% 245 ± 2% iostat.sdc.w/s
1020 ± 2% +21.7% 1242 ± 2% vmstat.system.in
200 ± 0% +22.1% 245 ± 2% iostat.sda.w/s
200 ± 0% +22.2% 245 ± 2% iostat.sdb.w/s
69.99 ± 0% -12.4% 61.34 ± 2% turbostat.PkgWatt
3943 ± 2% -8.9% 3593 ± 1% vmstat.system.cs
1.51 ± 1% +6.1% 1.60 ± 2% iostat.sdb.avgqu-sz
3.21 ± 0% +5.4% 3.39 ± 1% turbostat.RAMWatt
256182 ± 1% -4.2% 245424 ± 1% iostat.md0.avgqu-sz
more detailed changes about ivb44/fsmark/1x-1t-3HDD-RAID5-f2fs-4M-120G-NoSync
---------
c1dfe87e41d9c292 d4b4c2cdffab86f5c7594c4463
---------------- --------------------------
%stddev %change %stddev
\ | \
37.00 ± 0% -15.9% 31.10 ± 0% fsmark.files_per_sec
63414 ± 4% +57.6% 99945 ± 1% fsmark.time.voluntary_context_switches
830 ± 0% +18.8% 987 ± 0% fsmark.time.elapsed_time
830 ± 0% +18.8% 987 ± 0% fsmark.time.elapsed_time.max
9 ± 0% -14.8% 7 ± 6% fsmark.time.percent_of_cpu_this_job_got
1.48 ± 20% +357.3% 6.75 ± 5% turbostat.Pkg%pc6
63414 ± 4% +57.6% 99945 ± 1% time.voluntary_context_switches
109 ± 15% -37.8% 68 ± 20% time.involuntary_context_switches
338 ± 17% +57.6% 533 ± 0% cpuidle.POLL.usage
2691 ± 1% -20.3% 2144 ± 12% proc-vmstat.kswapd_high_wmark_hit_quickly
1060792 ± 0% +20.2% 1275544 ± 0% cpuidle.C6-IVT.usage
3.876e+10 ± 0% +19.3% 4.625e+10 ± 0% cpuidle.C6-IVT.time
830 ± 0% +18.8% 987 ± 0% time.elapsed_time.max
830 ± 0% +18.8% 987 ± 0% time.elapsed_time
39984 ± 0% +18.6% 47434 ± 0% uptime.idle
856 ± 0% +18.4% 1014 ± 0% uptime.boot
15874 ± 12% +20.9% 19188 ± 6% slabinfo.anon_vma.active_objs
1942445 ± 0% +18.1% 2293524 ± 0% proc-vmstat.pgfault
15977 ± 12% +20.1% 19188 ± 6% slabinfo.anon_vma.num_objs
110388 ± 9% +13.0% 124724 ± 4% meminfo.DirectMap4k
3107 ± 8% -20.9% 2459 ± 15% numa-meminfo.node0.AnonHugePages
18408 ± 11% +15.0% 21165 ± 3% slabinfo.free_nid.active_objs
18880 ± 11% +13.7% 21465 ± 4% slabinfo.free_nid.num_objs
1125535 ± 0% -11.5% 996605 ± 1% cpuidle.C1-IVT.usage
9 ± 0% -14.8% 7 ± 6% time.percent_of_cpu_this_job_got
198260 ± 1% +11.7% 221366 ± 0% softirqs.SCHED
6.09 ± 2% -12.2% 5.34 ± 0% turbostat.CPU%c1
14203 ± 2% -13.1% 12346 ± 8% slabinfo.kmalloc-256.num_objs
13763 ± 3% -13.3% 11937 ± 9% slabinfo.kmalloc-256.active_objs
1255 ± 6% +10.1% 1383 ± 1% slabinfo.RAW.num_objs
1255 ± 6% +10.1% 1383 ± 1% slabinfo.RAW.active_objs
30.37 ± 3% +30.5% 39.62 ± 0% iostat.sdc.rrqm/s
31.23 ± 5% +28.0% 39.98 ± 1% iostat.sdb.rrqm/s
33.37 ± 3% +19.0% 39.72 ± 2% iostat.sda.rrqm/s
562 ± 0% -15.9% 472 ± 0% iostat.md0.w/s
17106 ± 0% -15.9% 14382 ± 0% iostat.sda.wrqm/s
17106 ± 0% -15.9% 14382 ± 0% iostat.sdc.wrqm/s
17106 ± 0% -15.9% 14382 ± 0% iostat.sdb.wrqm/s
69317 ± 0% -15.9% 58284 ± 0% iostat.sdc.wkB/s
69316 ± 0% -15.9% 58284 ± 0% iostat.sda.wkB/s
69317 ± 0% -15.9% 58284 ± 0% iostat.sdb.wkB/s
138603 ± 0% -15.9% 116543 ± 0% iostat.md0.wkB/s
138705 ± 0% -15.9% 116633 ± 0% vmstat.io.bo
213 ± 0% -14.5% 182 ± 0% iostat.sdb.w/s
213 ± 0% -14.5% 182 ± 0% iostat.sda.w/s
213 ± 0% -14.6% 182 ± 0% iostat.sdc.w/s
4731 ± 0% -12.7% 4131 ± 0% vmstat.system.cs
1133 ± 2% -12.3% 993 ± 0% vmstat.system.in
3.02 ± 3% -8.6% 2.76 ± 3% iostat.sdc.avgqu-sz
3.29 ± 2% -9.4% 2.98 ± 3% iostat.sdb.avgqu-sz
25 ± 19% -21.3% 19 ± 2% turbostat.Avg_MHz
3.10 ± 1% -9.4% 2.81 ± 1% iostat.sda.avgqu-sz
44.45 ± 1% -5.6% 41.94 ± 2% turbostat.CorWatt
0.75 ± 19% -20.1% 0.60 ± 4% turbostat.%Busy
74.92 ± 1% -4.9% 71.23 ± 2% turbostat.PkgWatt
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists