lucene-solr-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Krishna Raj <reach.krishna...@gmail.com>
Subject Boosting Query Time Complexity
Date Mon, 18 Mar 2019 22:32:18 GMT
Hello Experts,

My goal is to understand the time complexity of the Boosting Query as part
of a search in Solr:

sort=score+desc
defType=edismax
boost=<function>

I followed the stacktrace for the the search call and I am believe the time
complexity is as follows:

- Main Query time + multiple filters reduce the DocId set of lets say size
N.

- Boost Query iterates the N DocId set and scores using the function. Time
Complexity: O(N)

- Since Tim Sort is used worst case Time Complexity is O(NLogN).

Can I know is this is a correct understanding? *Or, does the Custom
function boost scoring happens in tandem with Document Matching ?*

*Call StackTrace:*

 com.<my_company>.functions.functionvalues.PmmlComputation.pmml(PmmlComputation.java:75)
 com.<my_company>.functions.functionvalues.PmmlDoubleValues.doubleVal(PmmlDoubleValues.java:23)
 org.apache.lucene.queries.function.docvalues.DoubleDocValues.floatVal(DoubleDocValues.java:49)
 org.apache.lucene.queries.function.valuesource.ProductFloatFunction.func(ProductFloatFunction.java:39)
 org.apache.lucene.queries.function.valuesource.MultiFloatFunction$1.floatVal(MultiFloatFunction.java:82)
 org.apache.lucene.queries.function.BoostedQuery$CustomScorer.score(BoostedQuery.java:129)
 org.apache.lucene.search.ConjunctionScorer.score(ConjunctionScorer.java:64)
 org.apache.lucene.search.TopScoreDocCollector$SimpleTopScoreDocCollector$1.collect(TopScoreDocCollector.java:64)
 org.apache.lucene.search.MultiCollector$MultiLeafCollector.collect(MultiCollector.java:174)
 org.apache.lucene.search.Weight$DefaultBulkScorer.scoreAll(Weight.java:221)
 org.apache.lucene.search.Weight$DefaultBulkScorer.score(Weight.java:172)
 org.apache.lucene.search.BulkScorer.score(BulkScorer.java:39)
 org.apache.lucene.search.IndexSearcher.search(IndexSearcher.java:669)
 org.apache.lucene.search.IndexSearcher.search(IndexSearcher.java:473)
 org.apache.solr.search.SolrIndexSearcher.buildAndRunCollectorChain(SolrIndexSearcher.java:211)
 org.apache.solr.search.SolrIndexSearcher.getDocListAndSetNC(SolrIndexSearcher.java:1861)
 org.apache.solr.search.SolrIndexSearcher.getDocListC(SolrIndexSearcher.java:1578)
 org.apache.solr.search.SolrIndexSearcher.search(SolrIndexSearcher.java:581)
 org.apache.solr.handler.component.QueryComponent.process(QueryComponent.java:528)
 org.apache.solr.handler.component.SearchHandler.handleRequestBody(SearchHandler.java:295)
 org.apache.solr.handler.RequestHandlerBase.handleRequest(RequestHandlerBase.java:154)
 org.apache.solr.core.SolrCore.execute(SolrCore.java:2089)
 org.apache.solr.servlet.HttpSolrCall.execute(HttpSolrCall.java:652)

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message