Large sync writes perform worse with slog

For synchronous write workloads with large IO sizes, a pool configured
with a slog performs worse than one with an embedded zil:

sequential_writes 1m sync ios, 16 threads
  Write IOPS:              1292          438   -66.10%
  Write Bandwidth:      1323570       448910   -66.08%
  Write Latency:       12128400     36330970      3.0x

sequential_writes 1m sync ios, 32 threads
  Write IOPS:              1293          430   -66.74%
  Write Bandwidth:      1324184       441188   -66.68%
  Write Latency:       24486278     74028536      3.0x

The reason is the `zil_slog_bulk` variable. In `zil_lwb_write_open`,
if a zil block is greater than 768K, the priority of the write is
downgraded from sync to async. Increasing the value allows greater
throughput. To select a value for this PR, I ran an fio workload with
the following values for `zil_slog_bulk`:

    zil_slog_bulk    KiB/s
    1048576         422132
    2097152         478935
    4194304         533645
    8388608         623031
    12582912        827158
    16777216       1038359
    25165824       1142210
    33554432       1211472
    50331648       1292847
    67108864       1308506
    100663296      1306821
    134217728      1304998

At 64M, the results with a slog are now improved to parity with an
embedded zil:

sequential_writes 1m sync ios, 16 threads
  Write IOPS:               438         1288      2.9x
  Write Bandwidth:       448910      1319062      2.9x
  Write Latency:       36330970     12163408   -66.52%

sequential_writes 1m sync ios, 32 threads
  Write IOPS:               430         1290      3.0x
  Write Bandwidth:       441188      1321693      3.0x
  Write Latency:       74028536     24519698   -66.88%

None of the other tests in the performance suite (run with a zil or
slog) had a significant change, including the random_write_zil tests,
which use multiple datasets.

Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tony Nguyen <tony.nguyen@delphix.com>
Signed-off-by: John Wren Kennedy <john.kennedy@delphix.com>
Closes #14378

diff --git a/man/man4/zfs.4 b/man/man4/zfs.4
index 71a3e67ee67e31e27fe76f49e128dd21251f5f58..5f89f6adf1e33816a397193d8e3e37a7e3f873a8 100644 (file)
--- a/man/man4/zfs.4
+++ b/man/man4/zfs.4
@@ -2172,7 +2172,7 @@ if a volatile out-of-order write cache is enabled.
 Disable intent logging replay.
 Can be disabled for recovery from corrupted ZIL.
 .
-.It Sy zil_slog_bulk Ns = Ns Sy 786432 Ns B Po 768 KiB Pc Pq u64
+.It Sy zil_slog_bulk Ns = Ns Sy 67108864 Ns B Po 64 MiB Pc Pq u64
 Limit SLOG write size per commit executed with synchronous priority.
 Any writes above that will be executed with lower (asynchronous) priority
 to limit potential SLOG device abuse by single active ZIL writer.

diff --git a/module/zfs/zil.c b/module/zfs/zil.c
index 18c6cbf028b30d334cfefb857c36d48c97e85df9..a11886136994b797cc3be737893702b3013dd572 100644 (file)
--- a/module/zfs/zil.c
+++ b/module/zfs/zil.c
@@ -145,7 +145,7 @@ static int zil_nocacheflush = 0;
  * Any writes above that will be executed with lower (asynchronous) priority
  * to limit potential SLOG device abuse by single active ZIL writer.
  */
-static uint64_t zil_slog_bulk = 768 * 1024;
+static uint64_t zil_slog_bulk = 64 * 1024 * 1024;
 
 static kmem_cache_t *zil_lwb_cache;
 static kmem_cache_t *zil_zcw_cache;