org.apache.lucene.search

Class QueryCachingPolicy.CacheOnLargeSegments

java.lang.Object

org.apache.lucene.search.QueryCachingPolicy.CacheOnLargeSegments

All Implemented Interfaces:

QueryCachingPolicy

Enclosing interface:

QueryCachingPolicy

public static class QueryCachingPolicy.CacheOnLargeSegments
extends Object
implements QueryCachingPolicy

A simple policy that only caches on the largest segments of an index. The reasoning is that these segments likely account for most of the execution time of queries and are also more likely to stay around longer than small segments, which makes them more interesting for caching.

Nested Class Summary

Nested classes/interfaces inherited from interface org.apache.lucene.search.QueryCachingPolicy

QueryCachingPolicy.CacheOnLargeSegments

Field Summary

Fields

Modifier and Type	Field and Description
static QueryCachingPolicy.CacheOnLargeSegments	DEFAULT QueryCachingPolicy.CacheOnLargeSegments instance that only caches on segments that account for more than 3% of the total index size.

Fields inherited from interface org.apache.lucene.search.QueryCachingPolicy

ALWAYS_CACHE

Constructor Summary

Constructors

Constructor and Description
QueryCachingPolicy.CacheOnLargeSegments(int minIndexSize, float minSizeRatio) Create a QueryCachingPolicy.CacheOnLargeSegments instance that only caches on a given segment if the total number of documents in the index is greater than minIndexSize and the number of documents in the segment divided by the total number of documents in the index is greater than or equal to minSizeRatio.

Method Summary

Methods

Modifier and Type	Method and Description
void	onUse(Query query) Callback that is called every time that a cached filter is used.
boolean	shouldCache(Query query, LeafReaderContext context) Whether the given DocIdSet should be cached on a given segment.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

DEFAULT

public static final QueryCachingPolicy.CacheOnLargeSegments DEFAULT

QueryCachingPolicy.CacheOnLargeSegments instance that only caches on segments that account for more than 3% of the total index size. This should guarantee that all segments from the upper tier will be cached while ensuring that at most 33 segments can make it to the cache (given that some implementations such as LRUQueryCache perform better when the number of cached segments is low).

Constructor Detail

QueryCachingPolicy.CacheOnLargeSegments

public QueryCachingPolicy.CacheOnLargeSegments(int minIndexSize,
                                       float minSizeRatio)

Create a QueryCachingPolicy.CacheOnLargeSegments instance that only caches on a given segment if the total number of documents in the index is greater than minIndexSize and the number of documents in the segment divided by the total number of documents in the index is greater than or equal to minSizeRatio.

Method Detail

onUse

public void onUse(Query query)

Description copied from interface: QueryCachingPolicy

Callback that is called every time that a cached filter is used. This is typically useful if the policy wants to track usage statistics in order to make decisions.

Specified by:

onUse in interface QueryCachingPolicy

shouldCache

public boolean shouldCache(Query query,
                  LeafReaderContext context)
                    throws IOException

Description copied from interface: QueryCachingPolicy

Whether the given DocIdSet should be cached on a given segment. This method will be called on each leaf context to know if the filter should be cached on this particular leaf. The filter cache will first attempt to load a DocIdSet from the cache. If it is not cached yet and this method returns true then a cache entry will be generated. Otherwise an uncached set will be returned.

Specified by:

shouldCache in interface QueryCachingPolicy

Throws:

IOException