diff options
| author | Richard Wareing <rwareing@fb.com> | 2014-08-14 19:17:35 -0700 |
|---|---|---|
| committer | Kevin Vigor <kvigor@fb.com> | 2016-12-27 14:29:36 -0800 |
| commit | e1b74337f214fc1b185980228d7f7691b33877c4 (patch) | |
| tree | 7de53b3c5c76f23fa94991410c2f7c57219fc8f5 /libglusterfs/src | |
| parent | 88ef24b83f49c7d670720d59832d4e0f09efbe78 (diff) | |
Make Halo calculate & use average latencies, not realtime
Summary:
- Realtime latencies in practice have far too much jitter under real
loading conditions, instead let's use a running average which will get
very "heavy" over time such that temp spikes in brick latency will not
affect halo decisions.
Test Plan: - Run prove tests
Reviewed By: mmckeen
Change-Id: I5ebf9bc93c67d9a226287796dd7ca5eeb7b1cfa5
Signed-off-by: Kevin Vigor <kvigor@fb.com>
Reviewed-on: http://review.gluster.org/16301
CentOS-regression: Gluster Build System <jenkins@build.gluster.org>
NetBSD-regression: NetBSD Build System <jenkins@build.gluster.org>
Smoke: Gluster Build System <jenkins@build.gluster.org>
Reviewed-by: Shreyas Siravara <sshreyas@fb.com>
Diffstat (limited to 'libglusterfs/src')
| -rw-r--r-- | libglusterfs/src/xlator.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/libglusterfs/src/xlator.h b/libglusterfs/src/xlator.h index 571dce2aded..2e04893c487 100644 --- a/libglusterfs/src/xlator.h +++ b/libglusterfs/src/xlator.h @@ -927,7 +927,7 @@ struct _xlator { gf_loglevel_t loglevel; /* Log level for translator */ - int64_t client_latency; + fop_latency_t client_latency; /* for latency measurement */ fop_latency_t latencies[GF_FOP_MAXVALUE]; |