Stop using cached components in the _analyze API #19827
Labels
Comments
nik9000
added
blocker
:Search/Analysis
This was referenced Aug 5, 2016
johtani
added a commit
to johtani/elasticsearch
that referenced
this issue
Aug 10, 2016
Stop calling tokenizer/tokenFilters/chaFilter method of IndexService Add some getAnalysisProvider methods Change SynonymTokenFilterFactory constructor Closes elastic#19827
johtani
added a commit
to johtani/elasticsearch
that referenced
this issue
Aug 12, 2016
Add javadoc some methods Closes elastic#19827
s1monw
added a commit
to s1monw/elasticsearch
that referenced
this issue
Sep 22, 2016
…ping Today we hold on to all possible tokenizers, tokenfilters etc. when we create an index service on a node. This was mainly done to allow the `_analyze` API to directly access all these primitve. We fixed this in elastic#19827 and can now get rid of the AnalysisService entirely and replace it with a simple map like class. This ensures we don't create a gazillion long living objects that are entirely useless since they are never used in most of the indices. Also those objects might consume a considerable amount of memory since they might load stopwords or synonyms etc. Closes elastic#19828
s1monw
added a commit
that referenced
this issue
Sep 23, 2016
…ping (#20627) Today we hold on to all possible tokenizers, tokenfilters etc. when we create an index service on a node. This was mainly done to allow the `_analyze` API to directly access all these primitive. We fixed this in #19827 and can now get rid of the AnalysisService entirely and replace it with a simple map like class. This ensures we don't create a gazillion long living objects that are entirely useless since they are never used in most of the indices. Also those objects might consume a considerable amount of memory since they might load stopwords or synonyms etc. Closes #19828
s1monw
added a commit
that referenced
this issue
Sep 23, 2016
…ping (#20627) Today we hold on to all possible tokenizers, tokenfilters etc. when we create an index service on a node. This was mainly done to allow the `_analyze` API to directly access all these primitive. We fixed this in #19827 and can now get rid of the AnalysisService entirely and replace it with a simple map like class. This ensures we don't create a gazillion long living objects that are entirely useless since they are never used in most of the indices. Also those objects might consume a considerable amount of memory since they might load stopwords or synonyms etc. Closes #19828
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
We'd like to simplify
AnalysisServiceand remove most of its members (#19814), maybe reducing it to just aMap<String, Analyzer>. To do that, we should remove calls totokenizer,charFilter, andtokenFilterfrom the_analyzeAPI, instead rebuilding these analyzers on the fly. This will make some calls to the_analyzeAPI slower but it'll reduce the per index heap overhead.The text was updated successfully, but these errors were encountered: