| Blogs | Classifieds | Downloads | FlashChat | Gallery | Googlemap | Invite Friends | Links | Projects | Reviews | Wiki |
| |||||||||
Welcome to the pSeries Tech Forums,
our free peer-based support site for administrators, engineers and architects working with IBM pSeries servers and software. You are currently viewing our site as a guest which gives you limited access to view most discussions, articles, tutorials and access our other free features. By joining our community you will be able to collaborate with administrators, engineers and architects charged with designing, delivering or maintaining IBM pSeries server environments. Founded by a recognized IBM pSeries consultant and IBM Redbook author, pSeries Tech Forums was developed with the single mission of bringing IBM pSeries professionals together into a single self-help community. Registration is fast, simple and absolutely free to all IT professionals with responsibility for or interest in IBM pSeries servers. We invite you to join our community today! If you have any problems with the registration process or your account login, please contact contact support. |
| Our Sponsors | |
| | |
| Want to advertise? | |
![]() |
| | LinkBack | Thread Tools |
|
#1
| ||||
| ||||
| We just installed a new Oracle 10 Rac environment using to P550 serves with 8 gig of Ram. We have two internal disks mirrored and the rest are IBM DS4800 SAN disks. We currently have 6 databases in the rac environment. Problem we are having is that everytime we run a copy command or zip command we start paging. Hdisk0 and Hdisk1 go up to 100%. I was wondering a few things. 1. Should paging space be mirrored ? 2. If anyone has had this problem with Rac environment and what vio parameters did they use to maybe better tune the environment. 3. Just wondering if we put the OS disk on the SAN would it make difference. Any help is appreciated Thanks |
|
#2
| ||||
| ||||
Hi 1- Yes, it should be mirrored. Some people just don't mirror and just in case of disaster they go and create and activate a second mirror in the surviving disk. I usually create just one mirror on hdisk0 and mirror to the second rootvg disk. Mirroring at AIX LVM level is costly so you have to tune your Virtual Memory Manager (VMM) subsystem in such a way paging is avoided. You need to config the following parameters: maxperm minperm maxclient You can see the current values by running: vmo -a | egrep "maxperm|minperm|maxclient" By default they are 80/20/80. SAP, Oracle, DB2, Sybase, Lotus Domino, and personal experience recommends the following values: maxperm = 08 minperm = 03 maxclient = 08 There're 2 new values (lru related) that changed these values. 2- Yes i have had these problems and solved the way proposed. Modifyng vMM parameters. Please check Oracle documentation. 3- Hmmm. I don't think so. Your goal should be less paging space as possible. UNIX servers running with more than 25% of paging have performance problems. Hope this helps
__________________ cd3lgad0p |
|
#3
| ||||
| ||||
Thanks for your help, I'm playing with the setting in my test envornment an it shows very promising results. I did talk with IBM they recommented setting the parameters: 20 5 20 But I didn't really like the results so I switched back to your setttings. Thanks for the help |
|
#4
| ||||
| ||||
I also think you need to add more memory and tune oracle there is a redbook on Oracle best practice for AIX Also make sure that system layout is for best performance Oracle should have its own disks , Do not make paging to big
__________________ To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts. Glad to help |
|
#5
| ||||
| ||||
I found that this seems to be only happening when we used RMAN writing from asm to a filesystem. I implemented CIO option on the filesystem and this seems to have fixed the problem. What I think is happening is that RMAN using a block size that is to small to write to the filesystem granted I don't know what rman is doing underneath the covers but I think it thinks that it is writing to a raw partition instead of a filesystem. When I did a dd command to the filesystem using a block size of 256 I got the same results. After implementing CIO everthing was fine. |
![]() |
| Bookmarks |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | |
| |