[Veritas-bu] NetBackup 6.5.4 backup speed to IBM LTO-4 w/ENCRYPTION
2009-08-05 13:43:33
Nothing else going on with the host, run all backup jobs to a disk staging
unit, then perform the duplicate -> to tape.
Speed to LTO-4 (w/encryption+AES256):
Fiber Card: QLogic Corp. ISP2432-based 4Gb Fibre Channel
The bytes [io/bi] in (reading) off of a RAID-6 with 20+ 15k RPM drives:
procs -----------memory---------- ---swap-- -----io---- --system--
-----cpu------
r b swpd free buff cache si so bi bo in cs us sy id wa
st
1 0 252 85524 21748 15879868 0 0 123000 24 3294 2513 0 1 98
0 0
1 0 252 88820 22080 15856380 0 0 130200 64 4100 3591 2 2 90
6 0
0 0 252 89368 22056 15858372 0 0 123000 0 3367 2462 1 2 97
1 0
0 0 252 85348 22020 15862384 0 0 118900 0 3353 2394 0 1 98
0 0
0 1 252 88444 22024 15858284 0 0 118936 708 3223 4899 1 1 97
1 0
0 1 252 89748 22036 15852700 0 0 124440 0 3546 2719 1 2 96
1 0
0 0 252 85780 22024 15859984 0 0 122436 80 3231 2465 0 1 98
0 0
0 0 252 86200 22028 15859896 0 0 114808 0 3139 2435 0 1 98
1 0
0 1 252 86220 22016 15856048 0 0 119468 8 3231 2985 0 1 98
1 0
0 0 252 86216 22008 15859656 0 0 126536 448 3247 2518 0 1 98
1 0
0 0 252 90416 22012 15855884 0 0 118900 0 3145 2457 0 1 98
0 0
0 1 252 87356 21904 15855712 0 0 94864 0 2705 2301 0 1 97
2 0
0 0 252 82968 21936 15863436 0 0 147036 0 3590 2605 0 2 98
1 0
0 0 252 83140 21948 15863620 0 0 114800 28 3118 2438 0 1 98
0 0
1 0 252 83260 21924 15863288 0 0 123000 96 3162 2486 0 1 98
0 0
0 0 252 87156 21920 15859444 0 0 127100 24 3365 2495 0 1 98
1 0
0 0 252 83212 21924 15863032 0 0 118900 0 3148 2424 0 1 98
0 0
0 0 252 90696 21788 15855716 0 0 123000 0 3079 2540 0 1 98
0 0
0 0 252 90692 21096 15857020 0 0 123000 0 3203 2436 0 1 98
1 0
0 0 252 83012 21020 15864816 0 0 118900 0 3248 2439 0 1 98
0 0
2 0 252 81224 21048 15861512 0 0 115260 24 3073 2616 1 2 98
0 0
Average is right around 120MB/s which is correct for LTO-4. This is when the
backup was duplicating compressed data.
Earlier, when it was duplicating data (word documents, etc, heavily
compressible files):
procs -----------memory---------- ---swap-- -----io---- --system--
-----cpu------
r b swpd free buff cache si so bi bo in cs us sy id wa
st
2 0 252 88876 24316 15872360 0 0 287000 0 6172 3813 0 3 95
2 0
0 0 252 89664 24392 15871136 0 0 221400 0 4975 3285 0 2 96
1 0
1 0 252 90292 24424 15871016 0 0 229600 0 5169 3377 0 3 96
2 0
0 1 252 84012 24364 15873812 0 0 312164 0 6400 4102 0 4 94
2 0
0 0 252 85564 24428 15875432 0 0 200336 180 4736 3104 0 2 96
1 0
2 0 252 89040 24448 15872076 0 0 266500 0 5641 3681 0 3 95
2 0
0 0 252 85960 21984 15877568 0 0 332100 0 7138 4240 0 4 93
3 0
1 0 252 83072 22048 15880620 0 0 336200 0 6912 4284 0 4 94
2 0
2 0 252 82452 22168 15878648 0 0 336204 0 6925 4262 0 4 93
3 0
1 1 252 85192 22228 15874792 0 0 332664 40 6830 4243 0 4 94
2 0
0 1 252 86200 22228 15873428 0 0 336200 0 7100 4223 0 4 93
2 0
0 0 252 89968 22296 15873180 0 0 253636 540 5394 3549 0 3 94
3 0
If the bandwidth is available (high speed drives) etc, its quite nice to see
nice xfer rates such as this.
After the duplication to tape is finished, I'll see if I can find the average
speed across the entire duplication if that is logged in the bptm logs.
Parameters used:
NUMBER_DATA_BUFFERS: 32
NUMBER_DATA_BUFFERS_RESTORE: 32
SIZE_DATA_BUFFERS: 262144
SIZE_DATA_BUFFERS_DISK: 262144
Justin.
_______________________________________________
Veritas-bu maillist - Veritas-bu AT mailman.eng.auburn DOT edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-bu
|
<Prev in Thread] |
Current Thread |
[Next in Thread> |
- [Veritas-bu] NetBackup 6.5.4 backup speed to IBM LTO-4 w/ENCRYPTION,
Justin Piszcz <=
|
|
|