aboutsummaryrefslogtreecommitdiffstats
path: root/man
diff options
context:
space:
mode:
authorAlexander Motin <[email protected]>2024-05-01 14:07:20 -0400
committerGitHub <[email protected]>2024-05-01 11:07:20 -0700
commit645b83307918085ab2f0e12618809e348635b34f (patch)
tree50ca528cdfced535d6e3d88999b71bda8d712f58 /man
parent8fd3a5d02f3f6bad9e8e65b6aded694eae222bf2 (diff)
Improve write issue taskqs utilization
- Reduce number of allocators on small system down to one per 4 CPU cores, keeping maximum at 4 on 16+ core systems. Small systems should not have the lock contention multiple allocators supposed to solve, while having several metaslabs open and modified each TXG is not free. - Reduce number of write issue taskqs down to one per 16 CPU cores and an integer fraction of number of allocators. On mid- sized systems, where multiple allocators already make sense, too many write issue taskqs may reduce write speed on single-file workloads, since single file is handled by only one taskq to reduce fragmentation. On large systems, that can actually benefit from many taskq's better IOPS, the bottleneck is less important, since in worst case there will be at least 16 cores to handle it. - Distribute dnodes between allocators (and taskqs) in a round- robin fashion instead of relying on sync taskqs to be balanced. The last is not guarantied and may depend on scheduling. - Remove io_wr_iss_tq from struct zio. io_allocator is enough. Reviewed-by: Brian Behlendorf <[email protected]> Signed-off-by: Alexander Motin <[email protected]> Sponsored by: iXsystems, Inc. Closes #16130
Diffstat (limited to 'man')
-rw-r--r--man/man4/zfs.425
1 files changed, 15 insertions, 10 deletions
diff --git a/man/man4/zfs.4 b/man/man4/zfs.4
index ef0385d42..5edd80659 100644
--- a/man/man4/zfs.4
+++ b/man/man4/zfs.4
@@ -525,10 +525,17 @@ most ZPL operations (e.g. write, create) will return
.
.It Sy spa_num_allocators Ns = Ns Sy 4 Pq int
Determines the number of block alloctators to use per spa instance.
-Capped by the number of actual CPUs in the system.
+Capped by the number of actual CPUs in the system via
+.Sy spa_cpus_per_allocator .
.Pp
Note that setting this value too high could result in performance
degredation and/or excess fragmentation.
+Set value only applies to pools imported/created after that.
+.
+.It Sy spa_cpus_per_allocator Ns = Ns Sy 4 Pq int
+Determines the minimum number of CPUs in a system for block alloctator
+per spa instance.
+Set value only applies to pools imported/created after that.
.
.It Sy spa_upgrade_errlog_limit Ns = Ns Sy 0 Pq uint
Limits the number of on-disk error log entries that will be converted to the
@@ -2339,21 +2346,19 @@ Set value only applies to pools imported/created after that.
.
.It Sy zio_taskq_batch_tpq Ns = Ns Sy 0 Pq uint
Number of worker threads per taskq.
-Lower values improve I/O ordering and CPU utilization,
-while higher reduces lock contention.
+Higher values improve I/O ordering and CPU utilization,
+while lower reduce lock contention.
+Set value only applies to pools imported/created after that.
.Pp
If
.Sy 0 ,
generate a system-dependent value close to 6 threads per taskq.
Set value only applies to pools imported/created after that.
.
-.It Sy zio_taskq_wr_iss_ncpus Ns = Ns Sy 0 Pq uint
-Determines the number of CPUs to run write issue taskqs.
-.Pp
-When 0 (the default), the value to use is computed internally
-as the number of actual CPUs in the system divided by the
-.Sy spa_num_allocators
-value.
+.It Sy zio_taskq_write_tpq Ns = Ns Sy 16 Pq uint
+Determines the minumum number of threads per write issue taskq.
+Higher values improve CPU utilization on high throughput,
+while lower reduce taskq locks contention on high IOPS.
Set value only applies to pools imported/created after that.
.
.It Sy zio_taskq_read Ns = Ns Sy fixed,1,8 null scale null Pq charp