From c478ce86226821c18e2ec007f2877873d326f747 Mon Sep 17 00:00:00 2001 From: Rishabh Maurya Date: Thu, 5 Oct 2023 09:08:08 -0700 Subject: [PATCH] Documentation for configurable merge policy (#5137) * Documentation for configurable merge policy Signed-off-by: Rishabh Maurya * substitute indices with indexes Signed-off-by: Rishabh Maurya * substitute lucene with Lucene Signed-off-by: Rishabh Maurya * Update _im-plugin/index-settings.md Co-authored-by: Heather Halter Signed-off-by: Rishabh Maurya * Update _api-reference/cluster-api/cluster-settings.md Co-authored-by: Heather Halter Signed-off-by: Rishabh Maurya * Update _im-plugin/index-settings.md Co-authored-by: Nathan Bower Signed-off-by: Rishabh Maurya * Update _api-reference/cluster-api/cluster-settings.md Co-authored-by: Nathan Bower Signed-off-by: Rishabh Maurya --------- Signed-off-by: Rishabh Maurya Co-authored-by: Heather Halter Co-authored-by: Nathan Bower --- _api-reference/cluster-api/cluster-settings.md | 3 ++- _im-plugin/index-settings.md | 3 ++- 2 files changed, 4 insertions(+), 2 deletions(-) diff --git a/_api-reference/cluster-api/cluster-settings.md b/_api-reference/cluster-api/cluster-settings.md index 4df6039ac8..095f4f82cc 100644 --- a/_api-reference/cluster-api/cluster-settings.md +++ b/_api-reference/cluster-api/cluster-settings.md @@ -114,6 +114,7 @@ The following request field parameters are compatible with the cluster API. | cluster.persistent_tasks.allocation.enable | String | Enables or disables allocation for persistent tasks:

`all` – Allows persistent tasks to be assigned to nodes.

`none` – No allocations are allowed for persistent tasks. This does not affect persistent tasks already running.

Default is `all`. | | cluster.persistent_tasks.allocation.recheck_interval | Time unit | The cluster manager automatically checks whether or not persistent tasks need to be assigned when the cluster state changes in a significant way. There are other factors, such as memory usage, that will affect whether or not persistent tasks are assigned to nodes but do not otherwise cause the cluster state to change. This setting defines how often assignment checks are performed in response to these factors. Default is `30 seconds`, with a minimum of `10 seconds` being required. | | remote_store.moving_average_window_size | Integer | The moving average window size used to calculate the rolling statistic values exposed through the [Remote Store Stats API]({{site.url}}{{site.baseurl}}/tuning-your-cluster/availability-and-recovery/remote-store/remote-store-stats-api/). Default is `20`. Minimum enforced is `5`. | +| indices.time_series_index.default_index_merge_policy | String | This setting allows you to specify the default merge policy for time-series indexes, particularly for those with an `@timestamp` field, such as data streams. The two available options are `tiered` (default) and `log_byte_size`. We recommend using `log_byte_size` for time-series indexes to enhance the performance of range queries with the `@timestamp` field. To override the merge policy on a per-index basis, you can use the `index.merge.policy` index setting. | #### Example request @@ -144,4 +145,4 @@ For more information about transient settings, persistent settings, and preceden }, "transient":{} } -``` \ No newline at end of file +``` diff --git a/_im-plugin/index-settings.md b/_im-plugin/index-settings.md index d105d89b88..a123256944 100644 --- a/_im-plugin/index-settings.md +++ b/_im-plugin/index-settings.md @@ -42,6 +42,7 @@ index.routing_partition_size | The number of shards a custom routing value can g index.soft_deletes.retention_lease.period | The maximum amount of time to retain a shard's history of operations. Default is `12h`. index.load_fixed_bitset_filters_eagerly | Whether OpenSearch should preload cached filters. Available options are `true` and `false`. Default is `true`. index.hidden | Whether the index should be hidden. Hidden indexes are not returned as part of queries that have wildcards. Available options are `true` and `false`. Default is `false`. +index.merge.policy | This setting controls the merge policy for the Lucene segments. The available options are `tiered` and `log_byte_size`. The default is `tiered`, but for time-series data, such as log events, we recommend that you use the `log_byte_size` merge policy, which can improve query performance when conducting range queries on the `@timestamp` field. We recommend that you not change the merge policy of an existing index. Instead, configure this setting when creating a new index. ## Updating a static index setting @@ -119,4 +120,4 @@ PUT /testindex/_settings ``` {% include copy-curl.html %} -For more information about updating settings, including supported query parameters, see [Update settings]({{site.url}}{{site.baseurl}}/api-reference/index-apis/update-settings/). \ No newline at end of file +For more information about updating settings, including supported query parameters, see [Update settings]({{site.url}}{{site.baseurl}}/api-reference/index-apis/update-settings/).