nvme-multipath: implement "queue-depth" iopolicy

The round-robin path selector is inefficient in cases where there is a
difference in latency between paths.  In the presence of one or more
high latency paths the round-robin selector continues to use the high
latency path equally. This results in a bias towards the highest latency
path and can cause a significant decrease in overall performance as IOs
pile on the highest latency path. This problem is acute with NVMe-oF
controllers.

The queue-depth path selector sends I/O down the path with the lowest
number of requests in its request queue. Paths with lower latency will
clear requests more quickly and have less requests queued compared to
higher latency paths. The goal of this path selector is to make more use
of lower latency paths which will bring down overall IO latency and
increase throughput and performance.

Signed-off-by: Thomas Song <[email protected]>
[emilne: commandeered patch developed by Thomas Song @ Pure Storage]
Co-developed-by: Ewan D. Milne <[email protected]>
Signed-off-by: Ewan D. Milne <[email protected]>
Co-developed-by: John Meneghini <[email protected]>
Signed-off-by: John Meneghini <[email protected]>
Link: https://lore.kernel.org/linux-nvme/[email protected]/
Tested-by: Marco Patalano <[email protected]>
Tested-by: Jyoti Rani <[email protected]>
Tested-by: John Meneghini <[email protected]>
Reviewed-by: Randy Jennings <[email protected]>
Reviewed-by: Hannes Reinecke <[email protected]>
Reviewed-by: Sagi Grimberg <[email protected]>
Reviewed-by: Chaitanya Kulkarni <[email protected]>
Reviewed-by: Christoph Hellwig <[email protected]>
Signed-off-by: Keith Busch <[email protected]>

0 files changed, 0 insertions, 0 deletions