%cached 使用率不高,应该如何调整?
$ onstatIncorrectly built binary which accesses errno or h_errno directly. Needs to be fixed.
Informix Dynamic Server 2000 Version 9.21.UC2 -- On-Line -- Up 55 days 02:46:03 -- 1332288 Kbytes
Userthreads
addressflags sessid user tty wait tout locks nreads nwrites
57466018 ---P--D 1 informix - 0 0 0 30369 307850
57466608 ---P--F 0 informix - 0 0 0 0 74471412
57466bf8 ---P--- 6 informix - 0 0 0 0 2689198
574671e8 ---P--B 7 informix - 0 0 0 297 6433
57467dc8 ---P--D 11 informix - 0 0 0 0 1
57471238 Y--P--- 121427 informix - 57af8e28 0 1 376005 0
574729f8 Y--P--- 15038262 root - 57c46b98 0 1 48 0
5747d038 Y--P--- 944823 informix - 57d507e0 0 1 455173 0
5748ed38 Y--P--- 15360909 root - 57c6af88 0 1 0 0
592eb588 Y--P--- 19140980 batch - 57ae6980 0 1 14 33
592fde68 Y--P--- 19131931 batch - 57af88f0 0 1 39 0
5930d7c8 Y--P--- 19329301 erp - 583041e8 0 1 5743 5782
12 active, 256 total, 242 maximum concurrent
Profile
dskreads pagreads bufreads %cached dskwrits pagwrits bufwrits %cached
1023543631 11837814 2358754585 56.61 92092433 19965662 1707130417 94.61
isamtotopen start read write rewritedelete commit rollbk
843024294 342644534 840013618 1217225356 525551608 173859470 11279025 28649602 2323
gp_readgp_write gp_rewrt gp_del gp_alloc gp_freegp_curs
0 0 0 0 0 0 0
ovlock ovuserthread ovbuff usercpusyscpu numckpts flushes
14 0 0 228666.09 33780.72 15954 31918
bufwaits lokwaits lockreqs deadlksdltoutsckpwaits compress seqscans
87753742 57436133917724837 1 0 17442 43469012 107195934
ixda-RAidx-RA da-RA RA-pgsused lchwaits
37094409 131402 734210032 77123108017398701 增大buffers的个数 LOCKS 1024000 # Maximum number of locks
BUFFERS 512000 # Maximum number of shared buffers
是这个参数吗?
已经算比较大的了,我们服务器的配置是2G的内存。 OLTP那个值意义不大,你物理内存多大?
怎么还用9.21? 物理内存2G 是不是你的操作大部分都是insert操作呀。才造成你的读命中率很低,而写的cache则很高 貌似都是HD IObuff共用率較小
页:
[1]