Messages in this thread | | | Subject | Re: [PATCH #repost] SCSI: megaraid, fix suspend/resume sections | From | Zan Lynx <> | Date | Fri, 02 May 2008 11:15:33 -0600 |
| |
On Thu, 2008-05-01 at 15:30 -0700, Andrew Morton wrote: [cut] > > CONFIG_HOTPLUG is only settable to 'n' if you're CONFIG_EMBEDDED which > > has a zero set intersection with the users of megaraid, so in practical > > terms, there's no actual box it could kill. > > who suspends and resumes servers?
I read an article not long ago about a guy running an AMD renderfarm. He was working on power management techniques, including suspending systems that weren't in use. It made a huge difference in power costs, apparently.
I believe suspend is also used in some virtual server farms where physical systems are resumed and virtuals are migrated to satisfy higher CPU demands.
I could also see using it combined with a load balancer or proxy to bring up backup web or database servers.
Compared to using networked/USB power bars or management controllers to bring systems up and down, suspend/resume provides a faster response. -- Zan Lynx <zlynx@acm.org> [unhandled content-type:application/pgp-signature] | |