Simon Willnauer 320a206352 Switch back to ConcurrentMergeScheduler ...

Load tests showed that SerialMS has problems to keep up with
the merges under high load. We should switch back to CMS
until we have a better story to balance merge
threads / efforts across shards on a single node.

Closes #5817

2014-04-15 16:42:23 +02:00

..

allocation.asciidoc

Add getAsRatio to Settings class, allow DiskThresholdDecider to take percentages

2014-04-04 13:19:35 -06:00

analysis.asciidoc

[DOCS] Multiple doc fixes

2014-03-07 14:24:58 +01:00

cache.asciidoc

[DOCS] remove extraneous ` from cache page

2014-04-02 16:07:00 -06:00

codec.asciidoc

[DOCS] Multiple doc fixes

2014-03-07 14:24:58 +01:00

fielddata.asciidoc

Added global ordinals (unique incremental numbering for terms) to fielddata.

2014-04-07 11:06:41 +07:00

mapper.asciidoc

[DOCS] Multiple doc fixes

2014-03-07 14:24:58 +01:00

merge.asciidoc

Switch back to ConcurrentMergeScheduler

2014-04-15 16:42:23 +02:00

similarity.asciidoc

add lucene language model similarities (Dirichlet & JelinekMercer)

2014-04-07 10:48:03 +02:00

slowlog.asciidoc

Add more anchor links to documentation

2013-09-30 13:13:16 -06:00

store.asciidoc

fix docs to reflect removal of byte buffer memory

2014-02-03 09:54:30 -05:00

translog.asciidoc

randomize flush interval so multiple shards won't flush at the sam time

2014-01-07 19:58:28 +01:00