[jira] [Updated] (LUCENE-8689) Boolean DocValues Codec Implementation

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

[jira] [Updated] (LUCENE-8689) Boolean DocValues Codec Implementation

JIRA jira@apache.org

     [ https://issues.apache.org/jira/browse/LUCENE-8689?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Ivan Mamontov updated LUCENE-8689:
----------------------------------
    Description:
To avoid issues where some products become available/unavailable at some point in time after being out-of-stock, e-commerce search system designers need to embed up-to-date information about inventory availability right into the search engines. Key requirement is to be able to accurately filter out unavailable products and use availability as one of ranking signals. However, keeping availability data up-to-date is a non-trivial task. Straightforward implementation based on a partial updates of Lucene documents causes Solr cache trashing with negatively affected query performance and resource utilization.
 As an alternative solution we can use DocValues and build-in in-place updates where field values can be independently updated without touching inverted index, and while filtering by DocValues is a bit slower, overall performance gain is better. However existing long based docValues are not sufficiently optimized for carrying boolean inventory availability data:
 * All DocValues queries are internally rewritten into org.apache.lucene.search.DocValuesNumbersQuery which is based on direct iteration over all column values and typically much slower than using TermsQuery.
 * On every commit/merge codec has to iterate over DocValues a couple times in order to choose ths best compression algorithm suitable for given data. As a result for 4K fields and 3M max doc merge takes more than 10 minutes

This issue is intended to solve these limitations via special bitwise doc values format that uses internal representation of org.apache.lucene.util.FixedBitSet in order to store indexed values and load them at search time as a simple long array without additional decoding. There are several reasons for this:
 * At index time encoding is super fast without superfluous iterations over all values to choose ths best compression algorithm suitable for given data.</li>
 * At query time decoding is also simple and fast, no GC pressure and extra steps
 * Internal representation allows to perform random access in constant time

Limitations are:
 * Does not support non boolean fields
 * Boolean fields must be represented as long values 1 for true and 0 for false
 * Current implementation does not support advanced bit set formats like org.apache.lucene.util.SparseFixedBitSet or org.apache.lucene.util.RoaringDocIdSet

In order to evaluate performance gain I've wrote a simple JMH based benchmark  [^SynteticDocValuesBench70.java]  which allows to estimate a relative cost of DF filters. This benchmark creates 2 000 000 documents with 5 boolean columns with different density, where 10, 35, 50, 60 and 90 is an amount of documents with value 1. Each method tries to enumerate over all values in synthetic store field in all available ways:
 * baseline – in almost all cases Solr uses FixedBitSet in filter cache to keep store availability. This test just iterates over all bits.
 * docValuesRaw – iterates over all values of DV column, the same code is used in "post filtering", sorting and faceting.
 * docValuesNumbersQuery – iterates over all values produced by query/filter store:1, actually there is the only query implementation for DV based fields - DocValuesNumbersQuery. This means that Lucene rewrites all term, range and filter queries for non indexed filed into this fallback implementation.
 * docValuesBooleanQuery – optimized variant of DocValuesNumbersQuery, which support only two values – 0/1

!results2.png!

  was:
To avoid issues where some products become available/unavailable at some point in time after being out-of-stock, e-commerce search system designers need to embed up-to-date information about inventory availability right into the search engines. Key requirement is to be able to accurately filter out unavailable products and use availability as one of ranking signals. However, keeping availability data up-to-date is a non-trivial task. Straightforward implementation based on a partial updates of Lucene documents causes Solr cache trashing with negatively affected query performance and resource utilization.
 As an alternative solution we can use DocValues and build-in in-place updates where field values can be independently updated without touching inverted index, and while filtering by DocValues is a bit slower, overall performance gain is better. However existing long based docValues are not sufficiently optimized for carrying boolean inventory availability data:
 * All DocValues queries are internally rewritten into org.apache.lucene.search.DocValuesNumbersQuery which is based on direct iteration over all column values and typically much slower than using TermsQuery.
 * On every commit/merge codec has to iterate over DocValues a couple times in order to choose ths best compression algorithm suitable for given data. As a result for 4K fields and 3M max doc merge takes more than 10 minutes

This issue is intended to solve these limitations via special bitwise doc values format that uses internal representation of org.apache.lucene.util.FixedBitSet in order to store indexed values and load them at search time as a simple long array without additional decoding. There are several reasons for this:
 * At index time encoding is super fast without superfluous iterations over all values to choose ths best compression algorithm suitable for given data.</li>
 * At query time decoding is also simple and fast, no GC pressure and extra steps
 * Internal representation allows to perform random access in constant time

Limitations are:
 * Does not support non boolean fields
 * Boolean fields must be represented as long values 1 for true and 0 for false
 * Current implementation does not support advanced bit set formats like org.apache.lucene.util.SparseFixedBitSet or org.apache.lucene.util.RoaringDocIdSet

In order to evaluate performance gain I've wrote a simple JMH based benchmark  [^SynteticDocValuesBench70.java]  which allows to estimate a relative cost of DF filters. This benchmark creates 2 000 000 documents with 5 boolean columns with different density, where 10, 35, 50, 60 and 90 is an amount of documents with value 1. Each method tries to enumerate over all values in synthetic store field in all available ways:
 * baseline – in almost all cases Solr uses FixedBitSet in filter cache to keep store availability. This test just iterates over all bits.
 * docValuesRaw – iterates over all values of DV column, the same code is used in "post filtering", sorting and faceting.
 * docValuesNumbersQuery – iterates over all values produced by query/filter store:1, actually there is the only query implementation for DV based fields - DocValuesNumbersQuery. This means that Lucene rewrites all term, range and filter queries for non indexed filed into this fallback implementation.
 * docValuesBooleanQuery – optimized variant of DocValuesNumbersQuery, which support only two values – 0/1

!results.png!


> Boolean DocValues Codec Implementation
> --------------------------------------
>
>                 Key: LUCENE-8689
>                 URL: https://issues.apache.org/jira/browse/LUCENE-8689
>             Project: Lucene - Core
>          Issue Type: Improvement
>          Components: core/codecs
>            Reporter: Ivan Mamontov
>            Priority: Minor
>              Labels: patch, performance
>         Attachments: LUCENE-8689.patch, SynteticDocValuesBench70.java, results.png, results2.png
>
>
> To avoid issues where some products become available/unavailable at some point in time after being out-of-stock, e-commerce search system designers need to embed up-to-date information about inventory availability right into the search engines. Key requirement is to be able to accurately filter out unavailable products and use availability as one of ranking signals. However, keeping availability data up-to-date is a non-trivial task. Straightforward implementation based on a partial updates of Lucene documents causes Solr cache trashing with negatively affected query performance and resource utilization.
>  As an alternative solution we can use DocValues and build-in in-place updates where field values can be independently updated without touching inverted index, and while filtering by DocValues is a bit slower, overall performance gain is better. However existing long based docValues are not sufficiently optimized for carrying boolean inventory availability data:
>  * All DocValues queries are internally rewritten into org.apache.lucene.search.DocValuesNumbersQuery which is based on direct iteration over all column values and typically much slower than using TermsQuery.
>  * On every commit/merge codec has to iterate over DocValues a couple times in order to choose ths best compression algorithm suitable for given data. As a result for 4K fields and 3M max doc merge takes more than 10 minutes
> This issue is intended to solve these limitations via special bitwise doc values format that uses internal representation of org.apache.lucene.util.FixedBitSet in order to store indexed values and load them at search time as a simple long array without additional decoding. There are several reasons for this:
>  * At index time encoding is super fast without superfluous iterations over all values to choose ths best compression algorithm suitable for given data.</li>
>  * At query time decoding is also simple and fast, no GC pressure and extra steps
>  * Internal representation allows to perform random access in constant time
> Limitations are:
>  * Does not support non boolean fields
>  * Boolean fields must be represented as long values 1 for true and 0 for false
>  * Current implementation does not support advanced bit set formats like org.apache.lucene.util.SparseFixedBitSet or org.apache.lucene.util.RoaringDocIdSet
> In order to evaluate performance gain I've wrote a simple JMH based benchmark  [^SynteticDocValuesBench70.java]  which allows to estimate a relative cost of DF filters. This benchmark creates 2 000 000 documents with 5 boolean columns with different density, where 10, 35, 50, 60 and 90 is an amount of documents with value 1. Each method tries to enumerate over all values in synthetic store field in all available ways:
>  * baseline – in almost all cases Solr uses FixedBitSet in filter cache to keep store availability. This test just iterates over all bits.
>  * docValuesRaw – iterates over all values of DV column, the same code is used in "post filtering", sorting and faceting.
>  * docValuesNumbersQuery – iterates over all values produced by query/filter store:1, actually there is the only query implementation for DV based fields - DocValuesNumbersQuery. This means that Lucene rewrites all term, range and filter queries for non indexed filed into this fallback implementation.
>  * docValuesBooleanQuery – optimized variant of DocValuesNumbersQuery, which support only two values – 0/1
> !results2.png!



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]