11-10-2014
50% was at the time when you were lucky to have more than 2GB RAM...
Plenty buffers is good- too much can be a pain... the best results is when dbc_max_pct does not exceed 2.5GB so in the present case that would be something like 8%
reducing dbc_min_pct makes sence when you have memory shortage - Not the case here but could be reduced to 4%..
The issue I see but you havent mentionned it happned yet is more :
" Why cant I load more programs (no more memory error message...) when I have far more RAM than what is actually used???"
This is more what could happen to you, why?
Because you are using pseudoswap... HP-UX ( at least 11.11 ) is a OS using Virtual memory, to make it simple it will load first in swap to "reserve" the space then load in memory, in case of need to swap the space already reserved is used so in practical you could have far more program running than actual RAM only system will be slow because using its swap...
Your case is your swap device is far less than actual RAM, with pswedo-swap was not activated you could actually only use at maximum memory to the size of your swap: 16GB when you have 24! so turning swapmem_on to 1 lets you use the excedent RAM as swap so you can use more than 16GB
But to be able in case of crisis ot use all the RAM at other things than swap, you would have to create a secind swap device for performace - of the same size of the one you have: 16 GB bring your system to have 32GB of virtual memory.
Having 2 of same size will bring the OS to use both simultaneously (like load balancing) if it needs to
About SGA buffers etc thinking to make them bigger will be better can lead to performance surprises...
I have been called by a bank years (more than 10...)ago when they had performance issues when changing to batch mode just like you I asked for a quick sysdef and how much RAM they had... I knew the type of disks (FWSCSI...) so I asked them if it was something like the system freezes so often say every1/4 1/2 hour for a couple of minutes...
They wanted to know how I knew... (a full buffer of 5 GB will take more than 4 minutes to sync with those disks - the bigger the buffer , the more time - and its not linear but more like exponential hehe)
I said we could tune the JFS but a good start was to compile a new kernel with new buffer values, then see how much it improved, it solved their issue...
As fo SGA I have seen SQL request 30% faster on systems with 50% less SGA -The same! To much buffers to scan when data is not there will slow the system when with less SGA it would already have gone reading the disks
You said you are on a SAN - The SAN normally has huge cache also, so why waste resource by using sequencially 3 huges buffers before actually going to access the disks?
10 More Discussions You Might Find Interesting
1. AIX
Hi am facing high cpu utilization on my sybase server.
I have P550
Number Of Processors: 4
Processor Clock Speed: 1656 MHz
CPU Type: 64-bit
Kernel Type: 32-bit
LPAR Info: 1 65-D837E
Memory Size: 7840 MB
in topas it shows
Name PID CPU% PgSp Owner
dataserv 565264 ... (1 Reply)
Discussion started by: vjm
1 Replies
2. Shell Programming and Scripting
Hi Unix Gurus i am somewhat new to unix scripting so need your help to
create a script as below.
# This script would find the process consuming memory beyond a certain #limit. if the meemory consumption is more than 100% for a period of 1
# minute for the specific process. the script would... (0 Replies)
Discussion started by: robinforlinux
0 Replies
3. AIX
Hi,
i have two oracle instances running on my server with 7.5G RAM but it is showing almost 99% memory is full as you can see bellow.
# svmon
size inuse free pin virtual
memory 1974272 1954270 20002 336595 1800652
pg space 4194304 ... (7 Replies)
Discussion started by: salmanucit
7 Replies
4. AIX
How to find the memory utilization of AIX server using svmon -G output.
Sample output for svmon -G command from my AIX test server,
size inuse free pin virtual
memory 4014080 3995443 18637 575916 1876393
pg space 1179648 ... (3 Replies)
Discussion started by: maruthu
3 Replies
5. Red Hat
Hello All,
We have a jave server running on both linux and Solaris environments.
On solaris,it consumes only 600-700MB whereas it goes upto 21G in Linux.
I am monitoring the memory consumption through top command.
Is this high memory consumption expected in Linux? (1 Reply)
Discussion started by: prasperl
1 Replies
6. Solaris
Hi Experts,
Our servers running Solaris 10 with SAP Application. The memory utilization always >90%, but the process on SAP is too less even nothing.
Why memory utilization on solaris always looks high?
I have statement about memory on solaris, is this true:
Memory in solaris is used for... (4 Replies)
Discussion started by: edydsuranta
4 Replies
7. Solaris
Hello Guys,
I have one Solaris server with high memory utilization >90%. As per checking, below is the output for memory usage.
bash-3.00# ps -efo pmem,uid,pid,ppid,pcpu,comm | sort -r
%MEM UID PID PPID %CPU COMMAND
1.7 29496 20668 1 0.0 /opt/app/iw-home/tools/java/bin/java
1.5... (1 Reply)
Discussion started by: myrpthidesis
1 Replies
8. Red Hat
Hi All,
I have a java process which is showing high virtual memory utilization in client server. But the same process is showing comparitively lesser virtual memory consumption. I understand that virtual memory shown is not of much importance for the general user in normal condition, my client... (2 Replies)
Discussion started by: mritusmoi
2 Replies
9. Red Hat
Hi,
i am new to linux/RHEL 6.0 and i have two questions.
1) How to get the CPU utilization and Memory Utilization of all Services running currently?
2) How to get the CPU utilization and Memory Utilization of all Applications running currently?
Please help me to find the script.
... (2 Replies)
Discussion started by: nossam
2 Replies
10. UNIX for Advanced & Expert Users
Hi Fellas,
Not sure how I can dig in even further but we notice that one of our DB servers is showing high Sys% CPU usage even though when I execute the following command :
I can see that postgres is the only one using the CPU. So if anyone can advise me what would be the best way to... (3 Replies)
Discussion started by: arizah
3 Replies
tmpfs(7FS) File Systems tmpfs(7FS)
NAME
tmpfs - memory based file system
SYNOPSIS
#include <sys/mount.h>
mount (special, directory, MS_DATA, "tmpfs", NULL, 0);
DESCRIPTION
tmpfs is a memory based file system which uses kernel resources relating to the VM system and page cache as a file system. Once mounted, a
tmpfs file system provides standard file operations and semantics. tmpfs is so named because files and directories are not preserved across
reboot or unmounts, all files residing on a tmpfs file system that is unmounted will be lost.
tmpfs file systems can be mounted with the command:
mount -F tmpfs swap directory
Alternatively, to mount a tmpfs file system on /tmp at multi-user startup time (maximizing possible performance improvements), add the
following line to /etc/vfstab:
swap -/tmp tmpfs - yes -
tmpfs is designed as a performance enhancement which is achieved by caching the writes to files residing on a tmpfs file system. Perfor-
mance improvements are most noticeable when a large number of short lived files are written and accessed on a tmpfs file system. Large com-
pilations with tmpfs mounted on /tmp are a good example of this.
Users of tmpfs should be aware of some constraints involved in mounting a tmpfs file system. The resources used by tmpfs are the same as
those used when commands are executed (for example, swap space allocation). This means that large sized tmpfs files can affect the amount
of space left over for programs to execute. Likewise, programs requiring large amounts of memory use up the space available to tmpfs. Users
running into this constraint (for example, running out of space on tmpfs) can allocate more swap space by using the swap(1M) command.
Another constraint is that the number of files available in a tmpfs file system is calculated based on the physical memory of the machine
and not the size of the swap device/partition. If you have too many files, tmpfs will print a warning message and you will be unable to
create new files. You cannot increase this limit by adding swap space.
Normal file system writes are scheduled to be written to a permanent storage medium along with all control information associated with the
file (for example, modification time, file permissions). tmpfs control information resides only in memory and never needs to be written to
permanent storage. File data remains in core until memory demands are sufficient to cause pages associated with tmpfs to be reused at which
time they are copied out to swap.
An additional mount option can be specified to control the size of an individual tmpfs file system.
SEE ALSO
df(1M), mount(1M), mount_tmpfs(1M), swap(1M), mmap(2), mount(2), umount(2), vfstab(4)
System Administration Guide: Basic Administration
DIAGNOSTICS
If tmpfs runs out of space, one of the following messages will display in the console.
directory: File system full, swap space limit exceeded
This message appears because a page could not be allocated while writing to a file. This can occur if tmpfs is attempting to write more
than it is allowed, or if currently executing programs are using a lot of memory. To make more space available, remove unnecessary
files, exit from some programs, or allocate more swap space using swap(1M).
directory: File system full, memory allocation failed
tmpfs ran out of physical memory while attempting to create a new file or directory. Remove unnecessary files or directories or install
more physical memory.
WARNINGS
Files and directories on a tmpfs file system are not preserved across reboots or unmounts. Command scripts or programs which count on this
will not work as expected.
NOTES
Compilers do not necessarily use /tmp to write intermediate files therefore missing some significant performance benefits. This can be
remedied by setting the environment variable TMPDIR to /tmp. Compilers use the value in this environment variable as the name of the direc-
tory to store intermediate files.
swap to a tmpfs file is not supported.
df(1M) output is of limited accuracy since a tmpfs file system size is not static and the space available to tmpfs is dependent on the swap
space demands of the entire system.
SunOS 5.10 9 Oct 1990 tmpfs(7FS)