Blk-mq Is Almost Feature Complete & Fast With Linux 3.16
Merged for the Linux 3.13 kernel was the multi-queue block layer allows for better SSD performance with reduced latency and by balancing I/O workload across multiple CPU cores and supporting multiple hardware queues. With the upcoming Linux 3.16 kernel, the "blk-mq" code is expected to be feature complete and deliver great performance.
Jens Axboe sent in the pull request today for the core block I/O bits for the Linux 3.16 merge window. The new Facebook employee mentions that lots of development effort has gone into this block multi-queue code in recent months and "we're heading to where 3.16 will be a feature complete and performant blk-mq." A SCSI multi-queue implementation (scsi-mq) is expected to come for the Linux 3.17 kernel.
This 3.16 core block I/O pull request has some prep/support patches for scsi-mq, API and code clean-ups for blk-mq, and various other enhancements. Those unfamiliar with the multi-queue block layer can find out earlier articles on blk-mq.
This latest pull request for the Linux 3.16 kernel can be found via lkml.iu.edu.
Jens Axboe sent in the pull request today for the core block I/O bits for the Linux 3.16 merge window. The new Facebook employee mentions that lots of development effort has gone into this block multi-queue code in recent months and "we're heading to where 3.16 will be a feature complete and performant blk-mq." A SCSI multi-queue implementation (scsi-mq) is expected to come for the Linux 3.17 kernel.
This 3.16 core block I/O pull request has some prep/support patches for scsi-mq, API and code clean-ups for blk-mq, and various other enhancements. Those unfamiliar with the multi-queue block layer can find out earlier articles on blk-mq.
This latest pull request for the Linux 3.16 kernel can be found via lkml.iu.edu.
15 Comments