Embed Size (px)
Citation preview
Approaching Join Indexyet another one join algorithm
Mikhail Khludnevprincipal engineer
PRIVILEGED AND CONFIDENTIAL
• Grid Dynamics is a Silicon Valley-based leading provider of scalable, next-generation
commerce technology solutions
• Record of outperformance with Tier 1 retail clients
• Fortune 1000 client relationships
About Me
● principal engineer at Grid Dynamics
● spoke at few last LuceneRevolutions
● contributed BlockJoin query parser for Solr - {!parent}
● blogged about it at http://blog.griddynamics.com/
● tried to fix threads at DataImportHandler
http://google.com/+MikhailKhludnev
You are expected to know
● how Lucene searches/filters
● how it counts facets
● that there are segments
● what is DocValues
● why to join
● RDBMS joins: nested loop join, sort-merge join and hash join.
I’m expected to know
● query-time join
● index-time join
● yet another one join
Lucene/Solr Is Strong
● searching
○ filtering
● analytics
○ facets
○ pivots
○ stats
Lucene/Solr Is Weak in
● robust joins
○ multiple entities
○ relations
Joins in General
Joins in General
Joins in General
PK=FK
Joins in General
PK=FK
Joins in General
PK=FK
children
Joins in General
1:M
parents
Executing Join Query
q
Executing Join Query
q
Executing Join Query
q
Executing Join Query
qfq
Executing Join Query
qfq
Join in General
parents ∩ join-relation ∩ children
Join in General
parents ∩ join-relation ∩ children
● input
● output
● enumeration order
JoinUtil
q
“25”
“17”
“17”
“25”
“25”
“56”
“56”
“56”
“25”
“4”
“61”
FK[doc#]
JoinUtil
q
FK[doc#]
“17”
“17”
“25”
“25”
“56”
“56”
“56”
“25”
“4”
“61”
“25”
“25”
“17”
...
JoinUtil
q
PK
“1” →△…
“17”→△“25”→△...
“25”
“17”
...
“25”
“17”
“17”
“25”
“25”
“56”
“56”
“56”
“25”
“4”
“61”
FK[doc#]
JoinUtil
q
“1” →△…
“17”→△“25”→△...
“25”
“17”
...
fq
“25”
“17”
“17”
“25”
“25”
“56”
“56”
“56”
“25”
“4”
“61”
FK[doc#]PK
Block Join
doc#
Block Join
doc#
Block Join
doc#1
0
0
1
0
0
1
0
0
1
0
Block Join
doc#1
0
0
1
0
0
1
0
0
1
0
q
Block Join
doc#1
0
0
1
0
0
1
0
0
1
0
qfq
Block Join
doc#1
0
0
1
0
0
1
0
0
1
0
qfq
Comparison
JoinUtil BlockJoin
searching slow fast
reindexing
Comparison
JoinUtil BlockJoin
searching slow fast
reindexing fast slow
Comparison
doc#
JoinUtil BlockJoin
searching slow fast
reindexing fast slow
Comparison
doc#
JoinUtil BlockJoin
searching slow fast
reindexing fast slow
Comparison
doc#
JoinUtil BlockJoin
searching slow fast
reindexing fast slow
Comparison
JoinUtil BlockJoin
searching slow < ? < fast
reindexing fast > ? > slow
Join Index
q
doc#[doc#]
3
6
0
3
10
0
6
3
Join Index
q
doc#[doc#]
3
6
0
3
10
0
6
3
Join Index
q
doc#[doc#]
fq 3
6
0
3
10
0
6
3
Join Index
q
fq
doc#[doc#]
2
4
1
6
5 10
9
8
Join Index
q
doc#[doc#]
fq
2
4
1
6
5 10
9
8
Join Index
q
doc#[doc#]
fq
2
4
1
6
5 10
9
8
Benchmarking 2.9 M docs
https://github.com/m-khl/lucene-solr/tree/dvjoin-benchmark
Latency, ms
the bigger the worse
10
27
14
BlockJoin
(i-time)
Join Index
JoinUtil
(q-time)
doc#[doc#]
Indexing is still a problem
2
4
3
6
1
0
3
10
0
6
3
5 10
9
8
Further Plans
● incremental join-index update
● perhaps just calculate and cache it
● or put to dedicated index
● join in both directions
● calculate optimal execution plan of segments enumeration
Summary
● Joins in General
● JoinUtil vs Block-join
● {!scorejoin } - SOLR-6234
● updatable DocValues
● opportunities for improving query-time joins:
○ eliminate term enum
○ choose lower cardinality side for enumeration
References
● Searching relational content with Lucene's BlockJoinQuery
http://blog.mikemccandless.com/
● Solr Experience: search parent-child relations. Part I
Solr block-join support
http://blog.griddynamics.com/
● https://wiki.apache.org/solr/Join
● http://www.slideshare.net/martijnvg/document-relations
● https://issues.apache.org/jira/browse/SOLR-6234 {!scorejoin }
● Updatable DocValues Under the Hood
http://shaierera.blogspot.com/
● Subject: How to openIfChanged the most recent merge?
Off scope
Joins’ Zoo in Lucene
True Joins
● query-time join
○ JoinUtil
○ {!join }
○ {!scorejoin } - SOLR-6234
● index-time join aka block-join
{!parent}
Joins’ Zoo in Lucene
Workarounds
● term positions/SpanQueries
● FieldCollapsing/Grouping
● term decoration
○ spatial
● multivalue fields
True Joins
● query-time join
○ JoinUtil
○ {!join },
○ {!scorejoin } - SOLR-6234
● index-time join aka block-join
{!parent}
Joins’ Zoo in Lucene
Workarounds
● term positions/SpanQueries
● FieldCollapsing/Grouping
● term decoration
○ spatial
● multivalue fields
True Joins
● query-time join
○ JoinUtil
○ {!join },
○ {!scorejoin } - SOLR-6234
● index-time join aka block-join
{!parent}
JoinUtil
● query-time
● indexing is fast
● searching is slow, why?
○ expensive term enum
○ single enumeration order
BlockJoin
● index-time
● reindexing whole block is as expensive as mandatory
● searching is darn fast, however
○ can’t reorder child docs
