Processor Affinity

On 18th Aug 1998 kagel@bloomberg.net (Art S. Kagel) wrote:- There can be great benefit. Sometimes, on some systems (especially Numa architecture systems), there are performance gains above those achieved by the ONINIT affinity parameters to be had be affining the VPs yourself. I use a set of scripts that run onstat -g glo to a file and scans the report for the pids of VPs one VP class at a time affining each class, round robin, across the available processors. Doing this and switching back to AIO VPs we achieved 50% better performance than with ONINIT affinity and KAIO threads which Informix reports is 50% faster than ONINIT affinity and AIO VPs. One trick is to affine AIO VPs to the CPUs in the opposite order as the CPU VPs (ie affine CPU VP#1 to CPU#1 but AIO VP#1 to CPU #12 etc). Since the lowest numbered VPs of each class do most of the work this keeps the busiest CPU VPs from interfering with the busiest AIO VPs. My question is how do you affine processors apart from using the onconfig parameter? I am using IDS 7.3.0.UC6 on Solaris 2.6. Thanks for your time, Perry _______________________________________________________ Send a cool gift with your E-Card http://www.bluemountain.com/giftcenter/

Hi, I'm almost sure our friend "Art" has a list for all Operating Systems. Otherwise you should read your OS-manuals. Best regards, Stefan Weideneder Perigrine White wrote: > > On 18th Aug 1998 kagel@bloomberg.net (Art S. Kagel) wrote:- > There can be great benefit. Sometimes, on some systems (especially Numa > architecture systems), there are performance gains above those achieved by > the ONINIT affinity parameters to be had be affining the VPs yourself. I use > a set of scripts that run onstat -g glo to a file and scans the report for > the pids of VPs one VP class at a time affining each class, round robin, > across the available processors. Doing this and switching back to AIO VPs we > achieved 50% better performance than with ONINIT affinity and KAIO threads > which Informix reports is 50% faster than ONINIT affinity and AIO VPs. One > trick is to affine AIO VPs to the CPUs in the opposite order as the CPU VPs > (ie affine CPU VP#1 to CPU#1 but AIO VP#1 to CPU #12 etc). Since the lowest > numbered VPs of each class do most of the work this keeps the busiest CPU > VPs from interfering with the busiest AIO VPs. > > My question is how do you affine processors apart from using the onconfig > parameter? I am using IDS 7.3.0.UC6 on Solaris 2.6. > > Thanks for your time, > > Perry > > _______________________________________________________ > Send a cool gift with your E-Card > http://www.bluemountain.com/giftcenter/ -- Stefan Weideneder Phone: +49 89/3565478-2 --------------- --- Fax: +49 89/3565478-3 ------------- ------ mailto:/stefan at weideneder.de --- -------- http://www.weideneder.de -----

DG provides, if asked, an unsupported utility to set the affinity of a process to a CPU or CPU group. Sun likely has such from their advanced support group. Art S. Kagel Stefan Weideneder wrote: > > Hi, > > I'm almost sure our friend "Art" has a list for all > Operating Systems. > > Otherwise you should read your OS-manuals. > > Best regards, > > Stefan Weideneder > > Perigrine White wrote: > > > > On 18th Aug 1998 kagel@bloomberg.net (Art S. Kagel) wrote:- > > There can be great benefit. Sometimes, on some systems (especially Numa > > architecture systems), there are performance gains above those achieved by > > the ONINIT affinity parameters to be had be affining the VPs yourself. I use > > a set of scripts that run onstat -g glo to a file and scans the report for > > the pids of VPs one VP class at a time affining each class, round robin, > > across the available processors. Doing this and switching back to AIO VPs we > > achieved 50% better performance than with ONINIT affinity and KAIO threads > > which Informix reports is 50% faster than ONINIT affinity and AIO VPs. One > > trick is to affine AIO VPs to the CPUs in the opposite order as the CPU VPs > > (ie affine CPU VP#1 to CPU#1 but AIO VP#1 to CPU #12 etc). Since the lowest > > numbered VPs of each class do most of the work this keeps the busiest CPU > > VPs from interfering with the busiest AIO VPs. > > > > My question is how do you affine processors apart from using the onconfig > > parameter? I am using IDS 7.3.0.UC6 on Solaris 2.6. > > > > Thanks for your time, > > > > Perry > > > > _______________________________________________________ > > Send a cool gift with your E-Card > > http://www.bluemountain.com/giftcenter/ > > -- > Stefan Weideneder > > Phone: +49 89/3565478-2 --------------- > --- Fax: +49 89/3565478-3 ------------- > ------ mailto:/stefan at weideneder.de --- > -------- http://www.weideneder.de -----