# How many MergeScheduler are there in cluster? **URL:** \ **Category:** Elasticsearch\ **Created:** [May 28, 2016, 1:00pm UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221 "2016-05-28T13:00:12Z")\ **Posts on this page:** 7\ **Page:** 1
**Author:** ![itplayer](https://avatars.discourse-cdn.com/v4/letter/i/d26b3c/32.png) [@itplayer](https://discuss.elastic.co/u/itplayer)\ **Post date:** [May 28, 2016, 1:00pm UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/1 "2016-05-28T13:00:12Z")
Hi, if this is a common question, please forgive me as a newbie of Elasticsearch. If I have a ES cluster with 2 nodes and totally I have 10 indexes, each index with 4 shards. How many MergeScheduler are there in clusers? That's said a shard is actually a LuceneIndex, does it mean we have 40 MergeScheduler prepared when booting Elasticsearch server? Or each node has only one MergeScheduler? ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [May 30, 2016, 12:19am UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/2 "2016-05-30T00:19:49Z")
Have a look at [https://www.elastic.co/guide/en/elasticsearch/reference/2.3/index-modules-merge.html](https://www.elastic.co/guide/en/elasticsearch/reference/2.3/index-modules-merge.html) ---
**Author:** ![itplayer](https://avatars.discourse-cdn.com/v4/letter/i/d26b3c/32.png) [@itplayer](https://discuss.elastic.co/u/itplayer)\ **Post date:** [May 30, 2016, 6:03am UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/3 "2016-05-30T06:03:37Z")
I did read this article, but what I like to make sure is, how many "MergeScheduler instance" live in runtime? If MergeScheduler instance is per shard created and live in runtime, that I can expect setting different merge policy and parameters for each index independently. The same question for IndexWriter. I know each shard is actually a lucene files, but does Elasticsearch controll these shards(files) by using same instance(object) or place each instance per shards? ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [May 30, 2016, 6:06am UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/4 "2016-05-30T06:06:48Z")
> [@itplayer](#): > > I can expect setting different merge policy and parameters for each index independently There is a single merge policy for all indices, and it adapts to things. ---
**Author:** ![itplayer](https://avatars.discourse-cdn.com/v4/letter/i/d26b3c/32.png) [@itplayer](https://discuss.elastic.co/u/itplayer)\ **Post date:** [May 30, 2016, 6:33am UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/5 "2016-05-30T06:33:28Z")
Did you mean that I _can't_ set different merge policy parameters (default is TieredMergePolicy) for each index? We can only run _one_ setting for all index? ---
**Author:** ![mikemccand](https://avatars.discourse-cdn.com/v4/letter/m/f04885/32.png) [@mikemccand](https://discuss.elastic.co/u/mikemccand)\ **Post date:** [May 31, 2016, 12:22am UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/6 "2016-05-31T00:22:21Z")
Each index has its own merge settings, and every shard of every index has its own instance of e.g. `TieredMergePolicy` and `ConcurrentMergeScheduler` since each is a single Lucene index. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 5, 2017, 10:47pm UTC](https://discuss.elastic.co/t/how-many-mergescheduler-are-there-in-cluster/51221/7 "2017-07-05T22:47:54Z")