Tune for search speed

Elasticsearch heavily relies on the filesystem cache in order to make search fast. In general, you should make sure that at least half the available memory goes to the filesystem cache so that Elasticsearch can keep hot regions of the index in physical memory.

Search can cause a lot of randomized read I/O. When the underlying block device has a high readahead value, there may be a lot of unnecessary read I/O done, especially when files are accessed using memory mapping (see storage types).

Most Linux distributions use a sensible readahead value of 128KiB for a single plain device, however, when using software raid, LVM or dm-crypt the resulting block device (backing Elasticsearch path.data) may end up having a very large readahead value (in the range of several MiB). This usually results in severe page (filesystem) cache thrashing adversely affecting search (or update) performance.

You can check the current value in KiB using lsblk -o NAME,RA,MOUNTPOINT,TYPE,SIZE. Consult the documentation of your distribution on how to alter this value (for example with a udev rule to persist across reboots, or via blockdev --setra as a transient setting). We recommend a value of 128KiB for readahead.

If your searches are I/O-bound, consider increasing the size of the filesystem cache (see above) or using faster storage. Each search involves a mix of sequential and random reads across multiple files, and there may be many searches running concurrently on each shard, so SSD drives tend to perform better than spinning disks.

Directly-attached (local) storage generally performs better than remote storage because it is simpler to configure well and avoids communications overheads. With careful tuning it is sometimes possible to achieve acceptable performance using remote storage too. Benchmark your system with a realistic workload to determine the effects of any tuning parameters. If you cannot achieve the performance you expect, work with the vendor of your storage system to identify the problem.

If your searches are CPU-bound, consider using a larger number of faster CPUs.

Documents should be modeled so that search-time operations are as cheap as possible.

In particular, joins should be avoided. nested can make queries several times slower and parent-child relations can make queries hundreds of times slower. So if the same questions can be answered without joins by denormalizing documents, significant speedups can be expected.

The more fields a query_string or multi_match query targets, the slower it is. A common technique to improve search speed over multiple fields is to copy their values into a single field at index time, and then use this field at search time. This can be automated with the copy-to directive of mappings without having to change the source of documents. Here is an example of an index containing movies that optimizes queries that search over both the name and the plot of the movie by indexing both values into the name_and_plot field.

PUT movies
{
  "mappings": {
    "properties": {
      "name_and_plot": {
        "type": "text"
      },
      "name": {
        "type": "text",
        "copy_to": "name_and_plot"
      },
      "plot": {
        "type": "text",
        "copy_to": "name_and_plot"
      }
    }
  }
}

You should leverage patterns in your queries to optimize the way data is indexed. For instance, if all your documents have a price field and most queries run range aggregations on a fixed list of ranges, you could make this aggregation faster by pre-indexing the ranges into the index and using a terms aggregations.

For instance, if documents look like:

PUT index/_doc/1
{
  "designation": "spoon",
  "price": 13
}

and search requests look like:

GET index/_search
{
  "aggs": {
    "price_ranges": {
      "range": {
        "field": "price",
        "ranges": [
          { "to": 10 },
          { "from": 10, "to": 100 },
          { "from": 100 }
        ]
      }
    }
  }
}

Then documents could be enriched by a price_range field at index time, which should be mapped as a keyword:

PUT index
{
  "mappings": {
    "properties": {
      "price_range": {
        "type": "keyword"
      }
    }
  }
}

PUT index/_doc/1
{
  "designation": "spoon",
  "price": 13,
  "price_range": "10-100"
}

And then search requests could aggregate this new field rather than running a range aggregation on the price field.

GET index/_search
{
  "aggs": {
    "price_ranges": {
      "terms": {
        "field": "price_range"
      }
    }
  }
}

Not all numeric data should be mapped as a numeric field data type. Elasticsearch optimizes numeric fields, such as integer or long, for range queries. However, keyword fields are better for term and other term-level queries.

Identifiers, such as an ISBN or a product ID, are rarely used in range queries. However, they are often retrieved using term-level queries.

Consider mapping a numeric identifier as a keyword if:

If you’re unsure which to use, you can use a multi-field to map the data as both a keyword and a numeric data type.

If possible, avoid using script-based sorting, scripts in aggregations, and the script_score query. See Scripts, caching, and search speed.

Queries on date fields that use now are typically not cacheable since the range that is being matched changes all the time. However switching to a rounded date is often acceptable in terms of user experience, and has the benefit of making better use of the query cache.

For instance the below query:

PUT index/_doc/1
{
  "my_date": "2016-05-11T16:30:55.328Z"
}

GET index/_search
{
  "query": {
    "constant_score": {
      "filter": {
        "range": {
          "my_date": {
            "gte": "now-1h",
            "lte": "now"
          }
        }
      }
    }
  }
}

could be replaced with the following query:

GET index/_search
{
  "query": {
    "constant_score": {
      "filter": {
        "range": {
          "my_date": {
            "gte": "now-1h/m",
            "lte": "now/m"
          }
        }
      }
    }
  }
}

In that case we rounded to the minute, so if the current time is 16:31:29, the range query will match everything whose value of the my_date field is between 15:31:00 and 16:31:59. And if several users run a query that contains this range in the same minute, the query cache could help speed things up a bit. The longer the interval that is used for rounding, the more the query cache can help, but beware that too aggressive rounding might also hurt user experience.

GET index/_search
{
  "query": {
    "constant_score": {
      "filter": {
        "bool": {
          "should": [
            {
              "range": {
                "my_date": {
                  "gte": "now-1h",
                  "lte": "now-1h/m"
                }
              }
            },
            {
              "range": {
                "my_date": {
                  "gt": "now-1h/m",
                  "lt": "now/m"
                }
              }
            },
            {
              "range": {
                "my_date": {
                  "gte": "now/m",
                  "lte": "now"
                }
              }
            }
          ]
        }
      }
    }
  }
}

However such practice might make the query run slower in some cases since the overhead introduced by the bool query may defeat the savings from better leveraging the query cache.

Indices that are read-only may benefit from being merged down to a single segment. This is typically the case with time-based indices: only the index for the current time frame is getting new documents while older indices are read-only. Shards that have been force-merged into a single segment can use simpler and more efficient data structures to perform searches.

Global ordinals are a data structure that is used to optimize the performance of aggregations. They are calculated lazily and stored in the JVM heap as part of the field data cache. For fields that are heavily used for bucketing aggregations, you can tell Elasticsearch to construct and cache the global ordinals before requests are received. This should be done carefully because it will increase heap usage and can make refreshes take longer. The option can be updated dynamically on an existing mapping by setting the eager global ordinals mapping parameter:

PUT index
{
  "mappings": {
    "properties": {
      "foo": {
        "type": "keyword",
        "eager_global_ordinals": true
      }
    }
  }
}

If the machine running Elasticsearch is restarted, the filesystem cache will be empty, so it will take some time before the operating system loads hot regions of the index into memory so that search operations are fast. You can explicitly tell the operating system which files should be loaded into memory eagerly depending on the file extension using the index.store.preload setting.

Index sorting can be useful in order to make conjunctions faster at the cost of slightly slower indexing. Read more about it in the index sorting documentation.

There are multiple caches that can help with search performance, such as the filesystem cache, the request cache or the query cache. Yet all these caches are maintained at the node level, meaning that if you run the same request twice in a row, have 1 replica or more and use round-robin, the default routing algorithm, then those two requests will go to different shard copies, preventing node-level caches from helping.

Since it is common for users of a search application to run similar requests one after another, for instance in order to analyze a narrower subset of the index, using a preference value that identifies the current user or session could help optimize usage of the caches.

In addition to improving resiliency, replicas can help improve throughput. For instance if you have a single-shard index and three nodes, you will need to set the number of replicas to 2 in order to have 3 copies of your shard in total so that all nodes are utilized.

Now imagine that you have a 2-shards index and two nodes. In one case, the number of replicas is 0, meaning that each node holds a single shard. In the second case the number of replicas is 1, meaning that each node has two shards. Which setup is going to perform best in terms of search performance? Usually, the setup that has fewer shards per node in total will perform better. The reason for that is that it gives a greater share of the available filesystem cache to each shard, and the filesystem cache is probably Elasticsearch’s number 1 performance factor. At the same time, beware that a setup that does not have replicas is subject to failure in case of a single node failure, so there is a trade-off between throughput and availability.

So what is the right number of replicas? If you have a cluster that has num_nodes nodes, num_primaries primary shards in total and if you want to be able to cope with max_failures node failures at once at most, then the right number of replicas for you is max(max_failures, ceil(num_nodes / num_primaries) - 1).