A Proposal To Change The Default I/O Scheduler
A patch was volleyed into the Linux kernel development camp to change the default I/O scheduler for non-SATA disk drives.
Vivek Goyal, a Red Hat developer, questioned whether CFQ as the default scheduler in the Linux kernel is still the right choice. CFQ works well on a slow, rotational media like some Serial ATA disk drives, but under-performs for faster storage arrays, PCI-E solid-state drives, virtualized disks, etc. Therefore he's sent in a patch that would change the disk scheduler default for non-SATA drives to being the deadline scheduler rather than CFQ. Making deadline the default over CFQ for these faster storage mediums should provide a speed boost.
The Completely Fair Queuing scheduler has been the default since 2006 when it replaced the anticipatory scheduler as the mainline default.
From Vivek on the kernel mailing list, "One can argue that some SAS disks can be slow too and benefit from CFQ. Yes, but default IO scheduler choice is not perfect anyway. It just tries to cater to a wide variety of use cases out of the box. So I am throwing this patch out see if it flies. Personally, I think it might turn out to be a more reasonable default."
If other kernel developers agree, the default I/O scheduler could be changed for the Linux 3.5 kernel.
Vivek Goyal, a Red Hat developer, questioned whether CFQ as the default scheduler in the Linux kernel is still the right choice. CFQ works well on a slow, rotational media like some Serial ATA disk drives, but under-performs for faster storage arrays, PCI-E solid-state drives, virtualized disks, etc. Therefore he's sent in a patch that would change the disk scheduler default for non-SATA drives to being the deadline scheduler rather than CFQ. Making deadline the default over CFQ for these faster storage mediums should provide a speed boost.
The Completely Fair Queuing scheduler has been the default since 2006 when it replaced the anticipatory scheduler as the mainline default.
From Vivek on the kernel mailing list, "One can argue that some SAS disks can be slow too and benefit from CFQ. Yes, but default IO scheduler choice is not perfect anyway. It just tries to cater to a wide variety of use cases out of the box. So I am throwing this patch out see if it flies. Personally, I think it might turn out to be a more reasonable default."
If other kernel developers agree, the default I/O scheduler could be changed for the Linux 3.5 kernel.
9 Comments