Greetings,
I have a Alphaserver 8200 with 4-433mhz cpu's and 6 gig of ram. This
machine runs several different oracle databases. I have been modifying
/etc/sysconfigtab trying to improve performance using the white paper from
Compaq on large memory machines as an example.
I fear that my /etc/sysconfigtab is making things worse as a large
percentage of my time is spent in system mode and paging activity seems very
high.
Any comments/suggestions are most apreciated.
Thanks,
James Kurtenbach
Clinical Reference Laboratory
8433 Quivira
Lenexa, KS 66215
(913) 693 2508 (direct)
(913) 492 3652 (operator assisted)
kurtenbachj_at_crlcorp.com
Top shows:
load averages: 10.10, 10.64, 10.79
08:26:00
354 processes: 11 running, 11 waiting, 148 sleeping, 184 idle
CPU states: 46.7% user, 0.0% nice, 52.6% system, 0.6% idle
Memory: Real: 1325M/6029M act/tot Virtual: 17387M use/tot Free: 3495M
PID USERNAME PRI NICE SIZE RES STATE TIME CPU COMMAND
6570 oracle 48 0 1078M 51M run 41:04 28.30% oracle
27142 oracle 49 0 1078M 50M run 62:31 26.60% oracle
25284 applmgr 48 0 134M 17M run 4:02 24.30% oracle
29440 oracle 48 0 1078M 48M run 61:22 22.40% oracle
5216 oracle 47 0 1078M 48M run 45:51 22.00% oracle
12800 oracle 42 0 1078M 49M WAIT 51:57 21.70% oracle
25172 oracle 48 0 1078M 44M run 17:15 21.60% oracle
27377 oracle 47 0 1078M 49M run 63:19 20.90% oracle
25117 applmgr 47 0 134M 14M run 0:01 19.50% oracle
24949 applmgr 42 0 567M 164M sleep 2:15 17.20% oracle
20173 oracle 42 0 567M 146M WAIT 6:22 12.40% oracle
19022 oracle 42 0 567M 146M WAIT 6:40 11.40% oracle
13362 root 45 0 2824K 1679K run 0:00 9.80% ps
25163 oracle 44 0 17M 5529K sleep 4:06 6.90% dbsnmp
26534 oracle 42 0 1078M 46M WAIT 17:13 6.70% oracle
# vmstat -P
Total Physical Memory = 6144.00 M
= 786432 pages
Physical Memory Clusters:
start_pfn end_pfn type size_pages / size_bytes
0 256 pal 256 / 2.00M
256 786348 os 786092 / 6141.34M
786348 786432 pal 84 / 672.00k
Physical Memory Use:
start_pfn end_pfn type size_pages / size_bytes
256 276 unixtable 20 / 160.00k
276 284 scavenge 8 / 64.00k
284 768 text 484 / 3.78M
768 873 data 105 / 840.00k
873 1088 bss 215 / 1.68M
1088 1229 kdebug 141 / 1.10M
1229 1234 cfgmgmt 5 / 40.00k
1234 1235 locks 1 / 8.00k
1235 2378 unixtable 1143 / 8.93M
2378 2391 pmap 13 / 104.00k
2391 14957 vmtables 12566 / 98.17M
14957 786348 managed 771391 / 6026.49M
============================
Total Physical Memory Use: 786092 / 6141.34M
Managed Pages Break Down:
free pages = 445869
active pages = 55027
inactive pages = 105163
wired pages = 49837
ubc pages = 115771
==================
Total = 771667
WIRED Pages Break Down:
vm wired pages = 4095
ubc wired pages = 23766
meta data pages = 7860
malloc pages = 6614
contig pages = 134
user ptepages = 6888
kernel ptepages = 218
free ptepages = 5
==================
Total = 49580
Finally,
Here are the last 60 lines of sysconfigtab.
# Entries merged forward from /etc/sysconfigtab
generic:
msgbuf_size =16384
message-buffer-size =16384
proc:
per-proc-stack-size =8388608
per-proc-address-space =4398046511104
max-per-proc-address-space =4398046511104
max-proc-per-user =6000
maxusers=2048
ipc:
shm-max =2139095040
shm-mni =2048
shm-seg =512
sem-mni =2048
sem-msl =2048
sem-opm =400
sem-ume =400
ssm-threshold =0
vm:
vm-maxvas =10737418240
private-text =1
ubc-maxpercent=15
vm-ubcseqstartpercent =10
vm-page-free-target =2048
vm-mapentries =16384
vm-vpagemax =1310720
vfs:
bufcache =1
pts:
nptys =800
advfs:
AdvfsPreallocAccess =2528
AdvfsCacheMaxPercent=3
#
Received on Fri Mar 08 2002 - 15:00:16 NZDT