We currently migrate to MediaWiki from our old installation, but not all content has been migrated yet. Take a look at the Wiki Team page for instructions how to help or browse through our new wiki at wiki.linux-vserver.org to find the information already migrated.

Scheduling per Context

Talking about scheduling I would go for the O(1) implementation mentioned in "The CPU Scheduler", it looks clean. By easier/separate scheduling there would be scheduling of the schedulers? And isn't that too close to the UML solution?

- I would go for a O(1) scheduler, which does something like the token bucket thingy for each context ...

-- Asbjørn

Seperate Scheduling would just mean a new pair of run queues for each security context, as mentioned on the Scheduler page. This could satisfy "hard" scheduling requirements. However, it still does not approach the overheads of UML as we are still sharing all of the other kernel structures, such as the TCP/IP stack, memory management structures, etc.

-- SamV