query about benchmark

Hi, when I obtained the following information from the benchmark function of Informix OnLine XPS 7.1. +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ ********************************************************** c6p11sum60.lst = results of hwonstat script while running qsum60 on c6/PE11 ********************************************************** informix 6319 6277 0 16:46:40 ? 0:00 oninit -X6 informix 6337 6336 0 16:49:32 pts/0 0:00 grep -i oninit informix 6308 6277 0 16:46:32 ? 0:02 oninit -X6 informix 6277 1 0 16:45:47 ? 0:40 oninit -X6 informix 6318 6277 0 16:46:39 ? 0:01 oninit -X6 informix 6320 6277 0 16:46:41 ? 0:00 oninit -X6 informix 6303 6277 0 16:46:29 ? 0:02 oninit -X6 informix 6278 6277 0 16:45:51 ? 0:01 oninit -X6 informix 6310 6277 0 16:46:35 ? 0:01 oninit -X6 informix 6309 6277 0 16:46:34 ? 0:02 oninit -X6 informix 6317 6277 0 16:46:38 ? 0:00 oninit -X6 informix 6314 6277 0 16:46:36 ? 0:00 oninit -X6 informix 6321 6277 0 16:46:42 ? 0:00 oninit -X6 informix 6315 6277 0 16:46:37 ? 0:00 oninit -X6 informix 6322 6277 0 16:46:44 ? 0:01 oninit -X6 INFORMIX-OnLine Version 8.11.UF1 -- On-Line -- Up 00:03:41 -- 169160 Kbytes Profile dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached 28 38 152 81.58 14 14 5 0.00 isamtot open start read write rewrite delete commit rollbk 3 1 1 0 0 0 0 0 0 ovlock ovuserthread ovbuff usercpu syscpu numckpts flushes 0 0 0 12.53 36.57 1 2 bufwaits lokwaits lockreqs deadlks dltouts ckpwaits compress seqscans 0 0 0 0 0 0 0 0 ixda-RA idx-RA da-RA RA-pgsused lchwaits 0 0 0 0 2 MT global info: sessions threads vps lngspins 0 27 14 0 sched calls thread switches yield 0 yield n yield forever total: 359874 33654 326580 16917 155 per sec: 0 0 0 0 0 Virtual processor summary: class vps usercpu syscpu total cpu 1 5.78 34.01 39.79 aio 6 2.42 1.28 3.70 tli 1 0.40 0.11 0.51 lio 1 1.60 0.10 1.70 pio 1 1.61 0.06 1.67 adm 1 0.08 0.72 0.80 msc 1 0.25 0.14 0.39 fif 2 0.39 0.15 0.54 total 14 12.53 36.57 49.10 Individual virtual processors: vp pid class usercpu syscpu total 1 6277 cpu 5.78 34.01 39.79 2 6278 adm 0.08 0.72 0.80 3 6303 lio 1.60 0.10 1.70 4 6308 pio 1.61 0.06 1.67 5 6309 aio 1.40 0.26 1.66 6 6310 msc 0.25 0.14 0.39 7 6314 fif 0.20 0.07 0.27 8 6315 fif 0.19 0.08 0.27 9 6317 aio 0.20 0.26 0.46 10 6318 aio 0.21 0.55 0.76 11 6319 aio 0.19 0.04 0.23 12 6320 aio 0.22 0.09 0.31 13 6321 aio 0.20 0.08 0.28 14 6322 tli 0.40 0.11 0.51 tot 12.53 36.57 49.10 INFORMIX-OnLine Version 8.11.UF1 -- On-Line -- Up 00:03:47 -- 169160 Kbytes +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ the information of "aio 6 ...." is about the disk I/O cost, I believe. My question is that, the cost of reading data from disk to memory should including: 1. the time spends on DISK, "seek time, transfer time, rotational latency time"i" 2. the time spend on CPU, to put the data page in the cache,... 3. ... is the aio covers all of the above aspects, or it is the number 2 aspect(CPU cost) ? Many thanks. Jiang