DB : 11.2.0.4
OS solaris 5.10
Need help in understanding the memory utilization on the server. we have 200GB RAM allocated to the server, out of which 80GB is SGA_TARGET. constantly the memory is 93-95% (from the top command). We have 9 databases running on this server (the sum of 9 DB, SGA_TARGET is 80GB, biggest is SGA_TARGET=55GB), We still need to add 1 more database with 10GB SGA. but looks like won't be able to add until find how the memory is being used and if this can be freed. Here i am trying to understand where the remaining memory is being used.
Below helpful output for analysis. please let me know if need
1) Below from the top command:
Memory: 200G phys mem, 13G free mem, 256G total swap, 74G free swap
2) Allowed shareable memory is 100G for the oracle user form projects -l
project.max-shm-memory=(priv,107374182400,deny)
3) individual user memory from prstat -a. the sum of RSS is not even 100GB
NPROC USERNAME SWAP RSS MEMORY TIME CPU
2148 oracle 104G 98G 49% 473:27:11 5.0%
75 root 282M 223M 0.1% 64:54:20 0.1%
4 sapatrol 35M 47M 0.0% 23:27:00 0.0%
2 splunk 60M 89M 0.0% 2:54:35 0.0%
1 noaccess 96M 102M 0.0% 0:51:02 0.0%
2 uae9001 292K 1376K 0.0% 0:01:21 0.0%
4 SoysalSe 6920K 12M 0.0% 0:00:00 0.0%
2 saaddm 1768K 6440K 0.0% 0:00:01 0.0%
31 ChoudhAn 65M 44M 0.0% 0:00:13 0.0%
2 postfix 964K 2544K 0.0% 0:00:09 0.0%
6 daemon 8020K 6480K 0.0% 0:00:49 0.0%
Total: 2279 processes, 19649 lwps, load averages: 2.87, 3.17, 3.92
4) from the prstat -Z
$ prstat -Z
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
11471 oracle 2297M 2250M sleep 1 0 0:00:24 0.5% oracle/23
8676 oracle 55G 55G sleep 40 0 5:32:16 0.2% oracle/39
9405 oracle 6792K 6180K cpu17 9 0 0:00:03 0.2% prstat/1
6072 oracle 4324M 4279M sleep 59 0 0:00:04 0.2% oracle/11
4176 oracle 950M 853M sleep 59 0 24:14:44 0.1% java/177
9617 oracle 55G 55G sleep 1 0 0:00:01 0.1% oracle/1
.
. output truncated
.
17159 oracle 55G 55G sleep 45 0 0:03:00 0.0% oracle/18
6673 oracle 55G 55G sleep 1 0 9:56:12 0.0% oracle/1
ZONEID NPROC SWAP RSS MEMORY TIME CPU ZONE
----------------------------------------------------------------------------------------------------------------------------
3 2543 107G 100G 50% 562:02:36 2.7% <servername>
5) kstat -n system_pages|grep pp_kernel
pp_kernel 13478813
6) vmstat -S 3
kthr memory page disk faults cpu
r b w swap free si so pi po fr de sr lf s0 s1 s2 in sy cs us sy id
1 0 0 317758704 74279512 0 0 0 0 0 0 0 0 1 -0 24 17678 140834 44338 8 10 82
0 0 0 253910580 11620928 0 0 0 0 0 0 0 0 0 0 0 9537 96744 15407 12 8 80
1 0 0 253896648 11623116 0 0 0 0 0 0 0 0 0 0 0 11899 99664 31560 13 6 81
0 0 0 253821420 11588100 0 0 0 0 0 0 0 0 0 0 1 10062 123095 21509 14 7 79
1 0 0 253868856 11588856 0 0 0 0 0 0 0 0 0 0 39 10670 99003 21790 12 9 79
1 0 0 253849232 11577644 0 0 0 0 0 0 0 0 0 0 0 20817 135955 46667 15 6 79
1 0 0 253928564 11662480 0 0 0 0 0 0 0 0 0 0 3 10182 65995 18753 10 4 86
0 0 0 254074032 11822120 0 0 0 0 0 0 0 0 0 0 1 10041 109936 21480 11 5 84
1 0 0 254054344 11806036 0 0 0 0 0 0 0 0 0 0 1 16091 355762 37408 14 8 78
1 0 0 254052316 11809000 0 0 0 0 0 0 0 0 0 0 0 14363 48860 30588 9 3 87
0 0 0 254043464 11804212 0 0 0 0 0 0 0 0 0 0 0 17305 76425 36677 9 5 87
0 0 0 253752704 11787168 0 0 0 0 0 0 0 0 0 0 1 14696 98782 34416 11 7 82
6)
swap -l
swapfile dev swaplo blocks free
/dev/swap 4294967295,4294967295 8 536870904 154264736
Thank you very much