SCSI Multi-Queue Performance Appears Great For Linux 3.17
Written by Michael Larabel in Linux Kernel on 18 June 2014 at 06:22 AM EDT. 1 Comment
Building upon the major blk-mq work for the multi-queue block layer, the SCSI multi-queue code is now in good shape according to its developers, is delivering very promising performance results, and should be merged into the Linux 3.17 kernel cycle.

SCSI-mq is too late for Linux 3.16, but it was already anticipated it would come with Linux 3.17. SCSI-mq is about plugging the SCSI kernel code to take advantage of the multi-queue block layer code. "This patch adds support for an alternate I/O path in the scsi midlayer which uses the blk-mq infrastructure instead of the legacy request code. Use of blk-mq is fully transparent to drivers." The multi-queue block layer was first introduced in Linux 3.14 and is now fairly complete with Linux 3.16. The multi-queue block layer allows balancing I/O workload across multiple CPU cores and supports multiple hardware queues along with other performance optimizations.

Christoph Hellwig mentioned a few days ago, "At this point the code is ready for merging and use by developers and early adopters. The core blk-mq code isn't that suitable for slow devices yet, mostly due to the lack of an I/O scheduler, but Jens is working on it. Similarly there is no dm-multipath support for drivers using blk-mq yet, but I'm working on it. It should also be noted that the code doesn't actually support multiple hardware queues or fine grained tuning of the blk-mq parameters yet. All these could be added fairly easily as soon as low-level drivers want to make use of them."

In terms of the performance potential of this new code to better take advantage of multiple CPU cores and designed to work with (eventually) multiple hardware queues too, "The usage of blk-mq dramatically decreases CPU usage under all workloads going down from 100% CPU usage that the old setup can hit easily to usually less than 20% for maxing out storage subsystems with 512byte reads and writes, and it allows to easily archive millions of IOPS."

Bart Van Assche responded today with his performance figures for the new multi-queue code. "A very significant performance improvement for multithreaded workloads with use_blk_mq=Y. As an example, the number of I/O operations per second reported for the random write test increased with 170%. That means 2.7 times the performance of use_blk_mq=N." A small performance improvement was also noted for the traditional SCSI mid-layer code.

The Linux 3.16 kernel is still the better parts of two months out from being released when the Linux 3.17 kernel cycle will then begin with its final release not being expected until the late September/October time-frame, but at least this SCSI-mq code with significant performance improvements for enterprise Linux SCSI users should make the wait worthwhile. There's also other likely Linux 3.17 kernel changes we have already been covering on Phoronix.
About The Author
Author picture

Michael Larabel is the principal author of and founded the site in 2004 with a focus on enriching the Linux hardware experience. Michael has written more than 10,000 articles covering the state of Linux hardware support, Linux performance, graphics drivers, and other topics. Michael is also the lead developer of the Phoronix Test Suite, Phoromatic, and automated benchmarking software. He can be followed via Twitter or contacted via

Related Linux Kernel News
Popular News
Trending Reviews & Featured Articles