Look at the following top output, it is a classic 16 GB of RAM Linux server running Red Hat Enterprise Linux 4 64 bit.
I'm a little confused by the meaning of top's VIRT column value, as for most of the java processes (JBoss) that value is little higher than the RES value, which make sense for for 2 process is it very huge: 30.1g and 52.7g.
From what I was able to find the VIRT column is the total amount of virtual memory used by a process, meaning code, data, shared libs and swapped out pages. On that simple server, I only have 16 GB RAM and 2 GB swap (default value), how can I have those big numbers ?
When I do pmap on one of those process I get the same numbers and a lot of 262136K anon pages. I was wondering what could explain that situation?
Code:
top - 16:22:50 up 6:00, 2 users, load average: 0.30, 0.34, 0.34
Tasks: 136 total, 2 running, 134 sleeping, 0 stopped, 0 zombie
Cpu(s): 2.1% us, 2.7% sy, 0.0% ni, 95.3% id, 0.0% wa, 0.0% hi, 0.0% si
Mem: 16188436k total, 9753688k used, 6434748k free, 80992k buffers
Swap: 2096472k total, 0k used, 2096472k free, 1756028k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3341 jboss 16 0 53.7g 1.6g 61m S 1.8 10.1 21:23.21 java
3454 jboss 18 0 30.1g 1.4g 57m S 0.0 9.0 7:13.60 java
3571 jboss 20 0 3147m 2.3g 57m S 14.7 15.1 47:42.39 java
3641 jboss 25 0 1924m 1.0g 105m S 0.0 6.4 10:44.38 java
3512 jboss 16 0 1758m 791m 41m S 0.3 5.0 9:15.13 java
22703 jboss 25 0 1651m 634m 40m S 0.0 4.0 1:38.63 java
3869 jboss 17 0 1354m 86m 11m S 0.0 0.5 1:38.23 java
3077 nscd 16 0 188m 1984 1236 S 0.0 0.0 0:01.57 nscd
3265 root 16 0 110m 9.9m 5464 S 0.0 0.1 0:00.20 httpd
3306 apache 16 0 110m 5688 1012 S 0.0 0.0 0:00.00 httpd
3309 apache 16 0 110m 5684 1008 S 0.0 0.0 0:00.00 httpd
