[Bug 197516] TRIM on gmirror is slow and results in inresponsive system
bugzilla-noreply at freebsd.org
bugzilla-noreply at freebsd.org
Fri Feb 13 14:23:39 UTC 2015
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=197516
--- Comment #4 from Robert Schulze <rs at bytecamp.net> ---
Running the script from comment #3 on a gmirror with only one component, the
results are the following:
1) 9s
tty ada0 ada1 cpu
tin tout KB/t tps MB/s KB/t tps MB/s us ni sy in id
1 96 12.50 7 0.08 111.02 171 18.59 0 0 4 0 96
0 184 0.00 0 0.00 32.00 6613 206.65 0 0 18 0 82
0 59 0.00 0 0.00 32.00 3222 100.70 0 0 18 0 82
0 63 0.00 0 0.00 32.00 2817 88.02 0 0 18 0 82
0 62 0.00 0 0.00 32.00 2447 76.46 0 0 18 0 82
0 58 0.00 0 0.00 32.00 2634 82.31 0 0 16 0 84
0 58 0.00 0 0.00 32.00 3540 110.62 0 0 13 0 87
0 59 0.00 0 0.00 32.00 3682 115.06 0 0 13 0 87
0 59 0.00 0 0.00 32.00 5461 170.65 0 0 6 0 94
0 59 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
0 58 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
2) 5s
tty ada0 ada1 cpu
tin tout KB/t tps MB/s KB/t tps MB/s us ni sy in id
0 61 12.53 3 0.04 62.84 465 28.55 0 0 5 0 95
0 172 0.00 0 0.00 32.00 13636 426.09 0 0 21 0 79
0 60 0.00 0 0.00 32.00 8488 265.26 0 0 20 0 80
0 59 0.00 0 0.00 32.00 6232 194.73 0 0 13 0 87
0 59 0.00 0 0.00 32.00 1755 54.86 0 0 4 0 96
0 58 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
3) 9s
tty ada0 ada1 cpu
tin tout KB/t tps MB/s KB/t tps MB/s us ni sy in id
0 46 12.52 2 0.03 57.60 569 32.03 0 0 5 0 95
0 172 0.00 0 0.00 32.00 10313 322.27 0 0 22 0 78
0 60 0.00 0 0.00 32.00 3561 111.27 0 0 17 0 83
0 63 0.00 0 0.00 32.00 2442 76.32 0 0 18 0 82
0 62 0.00 0 0.00 32.00 2167 67.71 0 0 20 0 80
0 62 0.00 0 0.00 32.00 2090 65.32 0 0 20 0 80
0 61 0.00 0 0.00 32.00 2042 63.80 0 0 17 0 83
0 61 0.00 0 0.00 32.00 2368 74.00 0 0 14 0 86
0 61 0.00 0 0.00 32.00 5914 184.81 0 0 10 0 90
0 62 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
0 61 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
4) 14s
tty ada0 ada1 cpu
tin tout KB/t tps MB/s KB/t tps MB/s us ni sy in id
0 38 12.51 2 0.02 55.61 586 31.84 0 0 5 0 95
0 172 0.00 0 0.00 32.00 9241 288.79 0 0 19 0 81
0 63 0.00 0 0.00 32.00 3067 95.85 0 0 19 0 81
0 62 0.00 0 0.00 32.00 2305 72.03 0 0 19 0 81
0 62 0.00 0 0.00 32.00 1994 62.32 0 0 19 0 81
0 61 0.00 0 0.00 32.00 1784 55.75 0 0 18 0 82
0 60 0.00 0 0.00 32.00 1610 50.32 0 0 18 0 82
0 61 0.00 0 0.00 32.00 1446 45.20 0 0 18 0 82
0 62 0.00 0 0.00 32.00 1394 43.58 0 0 19 0 81
0 60 0.00 0 0.00 32.00 1800 56.26 0 0 12 0 88
0 61 0.00 0 0.00 32.00 1840 57.51 0 0 14 0 86
0 61 0.00 0 0.00 32.00 1900 59.37 0 0 12 0 88
0 61 0.00 0 0.00 32.00 1961 61.28 0 0 13 0 87
0 61 0.00 0 0.00 31.93 816 25.46 0 0 5 0 95
0 57 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
5) 7s
tty ada0 ada1 cpu
tin tout KB/t tps MB/s KB/t tps MB/s us ni sy in id
0 33 12.50 1 0.02 54.56 581 30.94 0 0 5 0 95
0 183 0.00 0 0.00 32.00 13082 408.80 0 0 23 0 76
0 60 0.00 0 0.00 32.00 4958 154.95 0 0 21 0 79
0 59 0.00 0 0.00 32.00 3581 111.92 0 0 19 0 81
0 59 0.00 0 0.00 32.00 3308 103.36 0 0 13 0 87
0 59 0.00 0 0.00 32.00 3331 104.10 0 0 13 0 87
0 59 0.00 0 0.00 31.98 2621 81.85 0 0 10 0 90
0 58 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
0 58 0.00 0 0.00 0.00 0 0.00 0 0 0 0 100
--
You are receiving this mail because:
You are the assignee for the bug.
More information about the freebsd-bugs
mailing list