mirror_zfs/module/zfs
ednadolski-ix 3bd4df3841
Improve ZFS objset sync parallelism
As part of transaction group commit, dsl_pool_sync() sequentially calls
dsl_dataset_sync() for each dirty dataset, which subsequently calls
dmu_objset_sync().  dmu_objset_sync() in turn uses up to 75% of CPU
cores to run sync_dnodes_task() in taskq threads to sync the dirty
dnodes (files).

There are two problems:

1. Each ZVOL in a pool is a separate dataset/objset having a single
   dnode.  This means the objsets are synchronized serially, which
   leads to a bottleneck of ~330K blocks written per second per pool.

2. In the case of multiple dirty dnodes/files on a dataset/objset on a
   big system they will be sync'd in parallel taskq threads. However,
   it is inefficient to to use 75% of CPU cores of a big system to do
   that, because of (a) bottlenecks on a single write issue taskq, and
   (b) allocation throttling.  In addition, if not for the allocation
   throttling sorting write requests by bookmarks (logical address),
   writes for different files may reach space allocators interleaved,
   leading to unwanted fragmentation.

The solution to both problems is to always sync no more and (if
possible) no fewer dnodes at the same time than there are allocators
the pool.

Reviewed-by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Signed-off-by: Edmund Nadolski <edmund.nadolski@ixsystems.com>
Closes #15197
2023-11-06 10:38:42 -08:00
..
abd.c RAIDZ: Use cache blocking during parity math 2023-10-30 14:54:27 -07:00
aggsum.c Remove bcopy(), bzero(), bcmp() 2022-03-15 15:13:42 -07:00
arc.c Unify arc_prune_async() code 2023-10-30 16:56:04 -07:00
blake3_zfs.c blake3: fix up bogus checksums in face of cpu migration 2023-05-01 17:21:27 -07:00
blkptr.c Remove bcopy(), bzero(), bcmp() 2022-03-15 15:13:42 -07:00
bplist.c Use list_remove_head() where possible. 2023-06-09 10:12:52 -07:00
bpobj.c Add explicit prefetches to bpobj_iterate(). 2023-07-21 11:50:48 -07:00
bptree.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
bqueue.c Batch enqueue/dequeue for bqueue 2023-01-10 13:39:22 -08:00
brt.c zfs_clone_range should return a descriptive error codes 2023-08-08 09:37:06 -07:00
btree.c btree: Implement faster binary search algorithm 2023-05-26 10:03:12 -07:00
dataset_kstats.c zil: Add some more statistics. 2023-05-25 13:51:53 -07:00
dbuf_stats.c Revert "Reduce dbuf_find() lock contention" 2022-09-22 12:59:41 -07:00
dbuf.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
ddt_zap.c Pack our DDT ZAPs a bit denser. 2023-06-30 09:42:02 -07:00
ddt.c ddt_addref: remove unnecessary phys fill when refcount is 0 2023-06-30 09:01:58 -07:00
dmu_diff.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
dmu_object.c zfs redact fails when dnodesize=auto 2023-02-16 09:23:39 -08:00
dmu_objset.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
dmu_recv.c Fix raw receive with different indirect block size. 2023-07-14 16:16:40 -07:00
dmu_redact.c Increase limit of redaction list by using spill block 2023-08-26 11:34:43 -07:00
dmu_send.c zdb: add -B option to generate backup stream 2023-06-05 11:54:42 -07:00
dmu_traverse.c Fix prefetching of indirect blocks while destroying 2023-03-24 10:20:07 -07:00
dmu_tx.c DMU: Do not pre-read holes during write 2023-10-11 16:37:21 -07:00
dmu_zfetch.c Add prefetch property 2023-10-24 11:00:07 -07:00
dmu.c ZIL: Avoid dbuf_read() before dmu_sync(). 2023-08-11 09:04:08 -07:00
dnode_sync.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
dnode.c Increase limit of redaction list by using spill block 2023-08-26 11:34:43 -07:00
dsl_bookmark.c Increase limit of redaction list by using spill block 2023-08-26 11:34:43 -07:00
dsl_crypt.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
dsl_dataset.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
dsl_deadlist.c Refactor dmu_prefetch(). 2023-08-07 13:54:41 -07:00
dsl_deleg.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
dsl_destroy.c Increase limit of redaction list by using spill block 2023-08-26 11:34:43 -07:00
dsl_dir.c Fix ENOSPC for extended quota 2023-09-28 14:10:07 -07:00
dsl_pool.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
dsl_prop.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
dsl_scan.c Do not request data L1 buffers on scan prefetch. 2023-07-20 09:10:04 -07:00
dsl_synctask.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
dsl_userhold.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
edonr_zfs.c Remove unused Edon-R variants 2023-03-14 15:59:58 -07:00
fm.c Use list_remove_head() where possible. 2023-06-09 10:12:52 -07:00
gzip.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
hkdf.c Remove bcopy(), bzero(), bcmp() 2022-03-15 15:13:42 -07:00
lz4_zfs.c Updated the lz4 decompressor 2022-01-07 10:36:49 -08:00
lz4.c lz4: Cherrypick fix for CVE-2021-3520 2022-01-12 16:14:36 -08:00
lzjb.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
metaslab.c Reduce number of metaslab preload taskq threads. 2023-10-06 09:04:00 -07:00
mmp.c Allow MMP to bypass waiting for other threads 2023-04-19 13:22:59 -07:00
multilist.c Cleanup: Specify unsignedness on things that should not be signed 2022-09-27 16:42:41 -07:00
objlist.c Implement Redacted Send/Receive 2019-06-19 09:48:12 -07:00
pathname.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
range_tree.c btree: Implement faster binary search algorithm 2023-05-26 10:03:12 -07:00
refcount.c Switch refcount tracking from lists to AVL-trees. 2023-06-14 08:02:27 -07:00
rrwlock.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
sa.c Cleanup: Remove constant comparisons reported by CodeQL 2023-03-08 13:51:46 -08:00
sha2_zfs.c Add generic implementation handling and SHA2 impl 2023-03-02 13:52:21 -08:00
skein_zfs.c Remove bcopy(), bzero(), bcmp() 2022-03-15 15:13:42 -07:00
spa_checkpoint.c Cleanup: 64-bit kernel module parameters should use fixed width types 2022-10-13 10:03:29 -07:00
spa_config.c Add mutex_enter_interruptible() for interruptible sleeping IOCTLs 2023-10-26 09:17:40 -07:00
spa_errlog.c Update the MOS directory on spa_upgrade_errlog() 2023-09-18 17:06:35 -07:00
spa_history.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
spa_log_spacemap.c Refactor dmu_prefetch(). 2023-08-07 13:54:41 -07:00
spa_misc.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
spa_stats.c Cleanup: Specify unsignedness on things that should not be signed 2022-09-27 16:42:41 -07:00
spa.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
space_map.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
space_reftree.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
THIRDPARTYLICENSE.cityhash OpenZFS 8484 - Implement aggregate sum and use for arc counters 2018-06-06 09:35:59 -07:00
THIRDPARTYLICENSE.cityhash.descrip OpenZFS 8484 - Implement aggregate sum and use for arc counters 2018-06-06 09:35:59 -07:00
txg.c Another set of vdev queue optimizations. 2023-06-27 09:09:48 -07:00
uberblock.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
unique.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_draid_rand.c Distributed Spare (dRAID) Feature 2020-11-13 13:51:51 -08:00
vdev_draid.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
vdev_indirect_births.c Remove bcopy(), bzero(), bcmp() 2022-03-15 15:13:42 -07:00
vdev_indirect_mapping.c Remove bcopy(), bzero(), bcmp() 2022-03-15 15:13:42 -07:00
vdev_indirect.c Don't emit cksum_{actual_expected} in ereport.fs.zfs.checksum events 2023-07-21 11:49:26 -07:00
vdev_initialize.c Add the ability to uninitialize 2023-05-18 10:02:20 -07:00
vdev_label.c Report ashift of L2ARC devices in zdb 2023-10-02 16:57:09 -07:00
vdev_mirror.c Improve too large physical ashift handling 2022-09-08 10:30:53 -07:00
vdev_missing.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_queue.c Another set of vdev queue optimizations. 2023-06-27 09:09:48 -07:00
vdev_raidz_math_aarch64_neon_common.h Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_aarch64_neon.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_aarch64_neonx2.c Fix Clang 15 compilation errors 2022-11-30 13:46:26 -08:00
vdev_raidz_math_avx2.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_avx512bw.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_avx512f.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_impl.h RAIDZ: Use cache blocking during parity math 2023-10-30 14:54:27 -07:00
vdev_raidz_math_powerpc_altivec_common.h Linux ppc64le ieee128 compat: Do not redefine __asm on external headers 2023-01-13 10:58:58 -08:00
vdev_raidz_math_powerpc_altivec.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_scalar.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_sse2.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math_ssse3.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_raidz_math.c Miscellaneous FreBSD compilation bugfixes 2023-04-06 10:35:02 -07:00
vdev_raidz.c Don't emit cksum_{actual_expected} in ereport.fs.zfs.checksum events 2023-07-21 11:49:26 -07:00
vdev_rebuild.c Tweak rebuild in-flight hard limit 2023-09-29 08:21:25 -07:00
vdev_removal.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
vdev_root.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
vdev_trim.c Zpool can start allocating from metaslab before TRIMs have completed 2023-10-12 11:01:54 -07:00
vdev.c Set spa_ccw_fail_time=0 when expanding a vdev. 2023-10-20 10:30:32 -07:00
zap_leaf.c Cleanup: Remove constant comparisons reported by CodeQL 2023-03-08 13:51:46 -08:00
zap_micro.c btree: Implement faster binary search algorithm 2023-05-26 10:03:12 -07:00
zap.c Cleanup: Use NULL when doing NULL pointer comparisons 2023-01-12 16:00:37 -08:00
zcp_get.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
zcp_global.c OpenZFS 8600 - ZFS channel programs - snapshot 2018-02-08 15:29:24 -08:00
zcp_iter.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
zcp_set.c Support setting user properties in a channel program 2020-02-14 13:41:42 -08:00
zcp_synctask.c Add zfs.sync.snapshot_rename 2022-09-02 13:31:19 -07:00
zcp.c nvpair: Constify string functions 2023-03-14 15:25:50 -07:00
zfeature.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
zfs_byteswap.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
zfs_chksum.c Add generic implementation handling and SHA2 impl 2023-03-02 13:52:21 -08:00
zfs_fm.c Don't emit cksum_{actual_expected} in ereport.fs.zfs.checksum events 2023-07-21 11:49:26 -07:00
zfs_fuid.c Use list_remove_head() where possible. 2023-06-09 10:12:52 -07:00
zfs_impl.c Add generic implementation handling and SHA2 impl 2023-03-02 13:52:21 -08:00
zfs_ioctl.c zvol: fix delayed update to block device ro entry 2023-10-31 09:50:38 -07:00
zfs_log.c ZIL: Cleanup sync and commit handling 2023-10-30 14:51:56 -07:00
zfs_onexit.c Use list_remove_head() where possible. 2023-06-09 10:12:52 -07:00
zfs_quota.c Revert "Do not persist user/group/project quota zap objects when unneeded" 2023-10-23 09:55:36 -07:00
zfs_ratelimit.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
zfs_replay.c Linux 6.3 compat: idmapped mount API changes 2023-04-10 14:15:36 -07:00
zfs_rlock.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
zfs_sa.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
zfs_vnops.c Fix block cloning between unencrypted and encrypted datasets 2023-10-31 13:49:41 -07:00
zil.c ZIO: Remove READY pipeline stage from root ZIOs 2023-10-25 15:22:25 -07:00
zio_checksum.c Don't emit cksum_{actual_expected} in ereport.fs.zfs.checksum events 2023-07-21 11:49:26 -07:00
zio_compress.c Skip memory allocation when compressing holes 2023-02-27 14:41:02 -08:00
zio_inject.c Cleanup: Switch to strlcpy from strncpy 2022-09-27 16:35:29 -07:00
zio.c Improve ZFS objset sync parallelism 2023-11-06 10:38:42 -08:00
zle.c Replace dead opensolaris.org license link 2022-07-11 14:16:13 -07:00
zrlock.c Micro-optimize zrl_remove() 2022-11-29 09:26:03 -08:00
zthr.c Switch from _Noreturn to __attribute__((noreturn)) 2022-03-23 08:51:00 -07:00
zvol.c zvol: fix delayed update to block device ro entry 2023-10-31 09:50:38 -07:00