# Function score filter function doesn't seem to be cached **URL:** \ **Category:** Elasticsearch\ **Created:** [February 12, 2015, 7:47pm UTC](https://discuss.elastic.co/t/function-score-filter-function-doesnt-seem-to-be-cached/22141 "2015-02-12T19:47:37Z")\ **Posts on this page:** 4\ **Page:** 1
**Author:** ![fforster](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/fforster/32/904_2.png) [@fforster](https://discuss.elastic.co/u/fforster)\ **Post date:** [February 12, 2015, 7:47pm UTC](https://discuss.elastic.co/t/function-score-filter-function-doesnt-seem-to-be-cached/22141/1 "2015-02-12T19:47:37Z")
I am using a function score with a function filter to score a subset of results higher. I specify to cache the function filter, but its performance is slower and slower the more clauses I add to the filter. A query with 2 geo\_polygons clauses returns relatively fast (30ms), but the more polygons clauses I add the slower it gets. With 50 polygons the call slows to 1s. Does elasticsearch actually cache the filter? Example query with 2 geo\_polygons clauses: { "query": { "function\_score": { "filter": { "bool": { "must": [ { "term": { "city": "LA" } } ] } }, "functions": [ { "filter": { "bool": { "should": [ { "geo\_polygon": { "location": { "points": [ [ -118.3971081, 34.081519 ], [ -118.3440351, 34.081519 ], [ -118.3440351, 34.131086 ], [ -118.3971081, 34.131086 ] ] } } }, { "geo\_polygon": { "location": { "points": [ [ -118.3907139, 34.05576 ], [ -118.361375, 34.05576 ], [ -118.361375, 34.083811 ], [ -118.3907139, 34.083811 ] ] } } } ], "\_cache": true } }, "boost\_factor": 2 } ], "score\_mode": "sum" } } } -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/c990d908-2b15-4797-9300-22a71aeb4972%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/c990d908-2b15-4797-9300-22a71aeb4972%40googlegroups.com). For more options, visit [https://groups.google.com/d/optout](https://groups.google.com/d/optout). ---
**Author:** ![Jason\_Wee](https://avatars.discourse-cdn.com/v4/letter/j/7ea924/32.png) [@Jason\_Wee](https://discuss.elastic.co/u/Jason_Wee)\ **Post date:** [February 13, 2015, 12:54am UTC](https://discuss.elastic.co/t/function-score-filter-function-doesnt-seem-to-be-cached/22141/2 "2015-02-13T00:54:27Z")
set explain to true and read the explain output if cache is use? Jason On Fri, Feb 13, 2015 at 3:47 AM, [fforster@homeaway.com](mailto:fforster@homeaway.com) wrote: > I am using a function score with a function filter to score a subset of > results higher. I specify to cache the function filter, but its performance > is slower and slower the more clauses I add to the filter. > > A query with 2 geo\_polygons clauses returns relatively fast (30ms), but > the more polygons clauses I add the slower it gets. With 50 polygons the > call slows to 1s. Does elasticsearch actually cache the filter? > > Example query with 2 geo\_polygons clauses: > { > "query": { > "function\_score": { > "filter": { > "bool": { > "must": [ > { > "term": { > "city": "LA" > } > } > ] > } > }, > "functions": [ > { > "filter": { > "bool": { > "should": [ > { > "geo\_polygon": { > "location": { > "points": [ > [ > -118.3971081, > 34.081519 > ], > [ > -118.3440351, > 34.081519 > ], > [ > -118.3440351, > 34.131086 > ], > [ > -118.3971081, > 34.131086 > ] > ] > } > } > }, > { > "geo\_polygon": { > "location": { > "points": [ > [ > -118.3907139, > 34.05576 > ], > [ > -118.361375, > 34.05576 > ], > [ > -118.361375, > 34.083811 > ], > [ > -118.3907139, > 34.083811 > ] > ] > } > } > } > ], > "\_cache": true > } > }, > "boost\_factor": 2 > } > ], > "score\_mode": "sum" > } > } > } > > -- > You received this message because you are subscribed to the Google Groups > "elasticsearch" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). > To view this discussion on the web visit > [https://groups.google.com/d/msgid/elasticsearch/c990d908-2b15-4797-9300-22a71aeb4972%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/c990d908-2b15-4797-9300-22a71aeb4972%40googlegroups.com) > [https://groups.google.com/d/msgid/elasticsearch/c990d908-2b15-4797-9300-22a71aeb4972%40googlegroups.com?utm\_medium=email&utm\_source=footer](https://groups.google.com/d/msgid/elasticsearch/c990d908-2b15-4797-9300-22a71aeb4972%40googlegroups.com?utm_medium=email&utm_source=footer) > . > For more options, visit [https://groups.google.com/d/optout](https://groups.google.com/d/optout). -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/CAHO4itze5wyDX%2Bp3OC%2BPpN%3Db1XHLQQvKb68ZEeOnAA9tEAwiBw%40mail.gmail.com](https://groups.google.com/d/msgid/elasticsearch/CAHO4itze5wyDX%2Bp3OC%2BPpN%3Db1XHLQQvKb68ZEeOnAA9tEAwiBw%40mail.gmail.com). For more options, visit [https://groups.google.com/d/optout](https://groups.google.com/d/optout). ---
**Author:** ![fforster](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/fforster/32/904_2.png) [@fforster](https://discuss.elastic.co/u/fforster)\ **Post date:** [February 13, 2015, 3:18pm UTC](https://discuss.elastic.co/t/function-score-filter-function-doesnt-seem-to-be-cached/22141/3 "2015-02-13T15:18:32Z")
Explain shows the cached filter, but based on the time it takes (took: 528) I doubt it's used efficiently { value: 2 description: Math.min of details: [ { value: 2 description: function score, score mode [sum] details: [ { value: 2 description: function score, product of: details: [ { value: 1 description: match filter: cache(BooleanFilter(+cache(olbValid:T) +BooleanFilter(GeoPolygonFilter(location, [[34.081519, -118.3971081], [34.081519, -118.3440351], [34.131086, -118.3440351], [34.131086, -118.3971081], [34.081519, -118.3971081]]) GeoPolygonFilter(location, [[34.05576, -118.3907139], [34.05576, -118.361375], [34.083811, -118.361375], [34.083811, -118.3907139], [34.05576, -118.3907139]]) GeoPolygonFilter(location, [[34.0412758, -118.528682], [34.0412758, -118.453384], [34.1324337, -118.453384], [34.1324337, -118.528682], [34.0412758, -118.528682]]) GeoPolygonFilter(location, [[34.0518801, -118.28436], [34.0518801, -118.2665471], [34.072782, -118.2665471], [34.072782, -118.28436], [34.0518801, -118.28436]]) GeoPolygonFilter(location, [[34.017233, -118.46146], [34.017233, -118.39865], [34.0532549, -118.39865], [34.0532549, -118.46146], [34.017233, -118.46146]]) GeoPolygonFilter(location, [[33.973535, -118.455804], [33.973535, -118.407918], [34.0268609, -118.407918], [34.0268609, -118.455804], [33.973535, -118.455804]]) GeoPolygonFilter(location, [[34.0491411, -118.4215721], [34.0491411, -118.406023], [34.0639819, -118.406023], [34.0639819, -118.4215721], [34.0491411, -118.4215721]]) GeoPolygonFilter(location, [[34.1036901, -118.3678789], [34.1036901, -118.3010089], [34.152483, -118.3010089], [34.152483, -118.3678789], [34.1036901, -118.3678789]]) GeoPolygonFilter(location, [[34.0790469, -118.396506], [34.0790469, -118.3669411], [34.1075319, -118.3669411], [34.1075319, -118.396506], [34.0790469, -118.396506]]) GeoPolygonFilter(location, [[34.048743, -118.259363], [34.048743, -118.250157], [34.056087, -118.250157], [34.056087, -118.259363], [34.048743, -118.259363]]) GeoPolygonFilter(location, [[34.0734518, -118.4822519], [34.0734518, -118.422647], [34.132908, -118.422647], [34.132908, -118.4822519], [34.0734518, -118.4822519]]) GeoPolygonFilter(location, [[33.9658801, -118.4404339], [33.9658801, -118.397569], [33.9839929, -118.397569], [33.9839929, -118.4404339], [33.9658801, -118.4404339]]) GeoPolygonFilter(location, [[34.065094, -118.4423781], [34.065094, -118.3867129], [34.1354771, -118.3867129], [34.1354771, -118.4423781], [34.065094, -118.4423781]]) GeoPolygonFilter(location, [[34.057085, -118.390748], [34.057085, -118.3720269], [34.072342, -118.3720269], [34.072342, -118.390748], [34.057085, -118.390748]]) GeoPolygonFilter(location, [[34.0541081, -118.4163189], [34.0541081, -118.389668], [34.069207, -118.389668], [34.069207, -118.4163189], [34.0541081, -118.4163189]]) GeoPolygonFilter(location, [[33.9755489, -118.419822], [33.9755489, -118.37006], [34.0173769, -118.37006], [34.0173769, -118.419822], [33.9755489, -118.419822]]) GeoPolygonFilter(location, [[33.9016261, -118.4371209], [33.9016261, -118.3735181], [33.931641, -118.3735181], [33.931641, -118.4371209], [33.9016261, -118.4371209]]) GeoPolygonFilter(location, [[33.8504661, -118.414124], [33.8504661, -118.384853], [33.878445, -118.384853], [33.878445, -118.414124], [33.8504661, -118.414124]]) GeoPolygonFilter(location, [[33.9955069, -119.0699391], [33.9955069, -118.557949], [34.1443899, -118.557949], [34.1443899, -119.0699391], [33.9955069, -119.0699391]]) GeoPolygonFilter(location, [[33.8728038, -118.430009], [33.8728038, -118.3785991], [33.9061261, -118.3785991], [33.9061261, -118.430009], [33.8728038, -118.430009]]) GeoPolygonFilter(location, [[34.024999, -118.5875051], [34.024999, -118.495177], [34.13165, -118.495177], [34.13165, -118.5875051], [34.024999, -118.5875051]]) GeoPolygonFilter(location, [[33.8041178, -118.405002], [33.8041178, -118.3703771], [33.8580759, -118.3703771], [33.8580759, -118.405002], [33.8041178, -118.405002]]) GeoPolygonFilter(location, [[34.041016, -118.651771], [34.041016, -118.5512071], [34.145413, -118.5512071], [34.145413, -118.651771], [34.041016, -118.651771]]) GeoPolygonFilter(location, [[33.9765188, -118.4904589], [33.9765188, -118.4443409], [34.0147499, -118.4443409], [34.0147499, -118.4904589], [33.9765188, -118.4904589]]) GeoPolygonFilter(location, [[33.95876, -118.471863], [33.95876, -118.4241309], [33.994206, -118.4241309], [33.994206, -118.471863], [33.95876, -118.471863]]) GeoPolygonFilter(location, [[33.9163209, -118.4640851], [33.9163209, -118.41719], [33.974274, -118.41719], [33.974274, -118.4640851], [33.9163209, -118.4640851]]) GeoPolygonFilter(location, [[33.993572, -118.508206], [33.993572, -118.483473], [34.0256331, -118.483473], [34.0256331, -118.508206], [33.993572, -118.508206]]) GeoPolygonFilter(location, [[34.018217, -118.521865], [34.018217, -118.4837659], [34.05056, -118.4837659], [34.05056, -118.521865], [34.018217, -118.521865]]) GeoPolygonFilter(location, [[34.0130348, -118.51226], [34.0130348, -118.470385], [34.046699, -118.470385], [34.046699, -118.51226], [34.0130348, -118.51226]]) GeoPolygonFilter(location, [[34.012045, -118.490271], [34.012045, -118.4528569], [34.0409041, -118.4528569], [34.0409041, -118.490271], [34.012045, -118.490271]]) GeoPolygonFilter(location, [[33.9954159, -118.4917131], [33.9954159, -118.4435171], [34.0278161, -118.4435171], [34.0278161, -118.4917131], [33.9954159, -118.4917131]]) GeoPolygonFilter(location, [[34.119054, -118.4225019], [34.119054, -118.35193], [34.157517, -118.35193], [34.157517, -118.4225019], [34.119054, -118.4225019]]) GeoPolygonFilter(location, [[34.063724, -118.4504281], [34.063724, -118.437812], [34.076625, -118.437812], [34.076625, -118.4504281], [34.063724, -118.4504281]])))) } { value: 2 description: static boost factor details: [ { value: 2 description: boostFactor } ] } ] } ] } On Thursday, February 12, 2015 at 1:47:37 PM UTC-6, [ffor...@homeaway.com](mailto:ffor...@homeaway.com) wrote: > I am using a function score with a function filter to score a subset of > results higher. I specify to cache the function filter, but its performance > is slower and slower the more clauses I add to the filter. > > A query with 2 geo\_polygons clauses returns relatively fast (30ms), but > the more polygons clauses I add the slower it gets. With 50 polygons the > call slows to 1s. Does elasticsearch actually cache the filter? > > Example query with 2 geo\_polygons clauses: > { > "query": { > "function\_score": { > "filter": { > "bool": { > "must": [ > { > "term": { > "city": "LA" > } > } > ] > } > }, > "functions": [ > { > "filter": { > "bool": { > "should": [ > { > "geo\_polygon": { > "location": { > "points": [ > [ > -118.3971081, > 34.081519 > ], > [ > -118.3440351, > 34.081519 > ], > [ > -118.3440351, > 34.131086 > ], > [ > -118.3971081, > 34.131086 > ] > ] > } > } > }, > { > "geo\_polygon": { > "location": { > "points": [ > [ > -118.3907139, > 34.05576 > ], > [ > -118.361375, > 34.05576 > ], > [ > -118.361375, > 34.083811 > ], > [ > -118.3907139, > 34.083811 > ] > ] > } > } > } > ], > "\_cache": true > } > }, > "boost\_factor": 2 > } > ], > "score\_mode": "sum" > } > } > } -- You received this message because you are subscribed to the Google Groups "elasticsearch" group. To unsubscribe from this group and stop receiving emails from it, send an email to [elasticsearch+unsubscribe@googlegroups.com](mailto:elasticsearch+unsubscribe@googlegroups.com). To view this discussion on the web visit [https://groups.google.com/d/msgid/elasticsearch/93cf1724-5871-4c45-a121-da049dfa184d%40googlegroups.com](https://groups.google.com/d/msgid/elasticsearch/93cf1724-5871-4c45-a121-da049dfa184d%40googlegroups.com). For more options, visit [https://groups.google.com/d/optout](https://groups.google.com/d/optout). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 12:32am UTC](https://discuss.elastic.co/t/function-score-filter-function-doesnt-seem-to-be-cached/22141/4 "2017-07-06T00:32:56Z")