This rewrites long sort as a `DistanceFeatureQuery`, which can
efficiently skip non-competitive blocks and segments of documents.
Depending on the dataset, the speedups can be 2 - 10 times.
The optimization can be disabled with setting the system property
`es.search.rewrite_sort` to `false`.
Optimization is skipped when an index has 50% or more data with
the same value.
Optimization is done through:
1. Rewriting sort as `DistanceFeatureQuery` which can
efficiently skip non-competitive blocks and segments of documents.
2. Sorting segments according to the primary numeric sort field(#44021)
This allows to skip non-competitive segments.
3. Using collector manager.
When we optimize sort, we sort segments by their min/max value.
As a collector expects to have segments in order,
we can not use a single collector for sorted segments.
We use collectorManager, where for every segment a dedicated collector
will be created.
4. Using Lucene's shared TopFieldCollector manager
This collector manager is able to exchange minimum competitive
score between collectors, which allows us to efficiently skip
the whole segments that don't contain competitive scores.
5. When index is force merged to a single segment, #48533 interleaving
old and new segments allows for this optimization as well,
as blocks with non-competitive docs can be skipped.
Closes#37043
Co-authored-by: Jim Ferenczi <jim.ferenczi@elastic.co>
* Optimize sort on numeric long and date fields (#39770)
Optimize sort on numeric long and date fields, when
the system property `es.search.long_sort_optimized` is true.
* Skip optimization if the index has duplicate data (#43121)
Skip sort optimization if the index has 50% or more data
with the same value.
When index has a lot of docs with the same value, sort
optimization doesn't make sense, as DistanceFeatureQuery
will produce same scores for these docs, and Lucene
will use the second sort to tie-break. This could be slower
than usual sorting.
* Sort leaves on search according to the primary numeric sort field (#44021)
This change pre-sort the index reader leaves (segment) prior to search
when the primary sort is a numeric field eligible to the distance feature
optimization. It also adds a tie breaker on `_doc` to the rewritten sort
in order to bypass the fact that leaves will be collected in a random order.
I ran this patch on the http_logs benchmark and the results are very promising:
```
| 50th percentile latency | desc_sort_timestamp | 220.706 | 136544 | 136324 | ms |
| 90th percentile latency | desc_sort_timestamp | 244.847 | 162084 | 161839 | ms |
| 99th percentile latency | desc_sort_timestamp | 316.627 | 172005 | 171688 | ms |
| 100th percentile latency | desc_sort_timestamp | 335.306 | 173325 | 172989 | ms |
| 50th percentile service time | desc_sort_timestamp | 218.369 | 1968.11 | 1749.74 | ms |
| 90th percentile service time | desc_sort_timestamp | 244.182 | 2447.2 | 2203.02 | ms |
| 99th percentile service time | desc_sort_timestamp | 313.176 | 2950.85 | 2637.67 | ms |
| 100th percentile service time | desc_sort_timestamp | 332.924 | 2959.38 | 2626.45 | ms |
| error rate | desc_sort_timestamp | 0 | 0 | 0 | % |
| Min Throughput | asc_sort_timestamp | 0.801824 | 0.800855 | -0.00097 | ops/s |
| Median Throughput | asc_sort_timestamp | 0.802595 | 0.801104 | -0.00149 | ops/s |
| Max Throughput | asc_sort_timestamp | 0.803282 | 0.801351 | -0.00193 | ops/s |
| 50th percentile latency | asc_sort_timestamp | 220.761 | 824.098 | 603.336 | ms |
| 90th percentile latency | asc_sort_timestamp | 251.741 | 853.984 | 602.243 | ms |
| 99th percentile latency | asc_sort_timestamp | 368.761 | 893.943 | 525.182 | ms |
| 100th percentile latency | asc_sort_timestamp | 431.042 | 908.85 | 477.808 | ms |
| 50th percentile service time | asc_sort_timestamp | 218.547 | 820.757 | 602.211 | ms |
| 90th percentile service time | asc_sort_timestamp | 249.578 | 849.886 | 600.308 | ms |
| 99th percentile service time | asc_sort_timestamp | 366.317 | 888.894 | 522.577 | ms |
| 100th percentile service time | asc_sort_timestamp | 430.952 | 908.401 | 477.45 | ms |
| error rate | asc_sort_timestamp | 0 | 0 | 0 | % |
```
So roughly 10x faster for the descending sort and 2-3x faster in the ascending case. Note
that I indexed the http_logs with a single client in order to simulate real time-based indices
where document are indexed in their timestamp order.
Relates #37043
* Remove nested collector in docs response
As we don't use cancellableCollector anymore, it should be removed from
the expected docs response.
* Use collector manager for search when necessary (#45829)
When we optimize sort, we sort segments by their min/max value.
As a collector expects to have segments in order,
we can not use a single collector for sorted segments.
Thus for such a case, we use collectorManager,
where for every segment a dedicated collector will be created.
* Use shared TopFieldCollector manager
Use shared TopFieldCollector manager for sort optimization.
This collector manager is able to exchange minimum competitive
score between collectors
* Correct calculation of avg value to avoid overflow
* Optimize calculating if index has duplicate data
All document scores are positive 32-bit floating point numbers. However, this
wasn't previously documented.
This can result in surprising behavior, such as precision loss, for users when
customizing scores using the function score query.
This commit updates an existing admonition in the function score query docs to
document the 32-bits precision limit. It also updates the search API reference
docs to note that `_score` is a 32-bit float.
Customers occasionally discover a known behavior in Elasticsearch's pagination that does not appear to be documented. This warning is intended to educate customers of this behavior while still highlighting alternative solutions.
* [DOCS] Add template docs to scripts. Reorder template examples.
* Adds a 'Search template' section to the 'How to use scripts' chapter.
This links to the 'Search template' chapter for detailed info and
examples.
* Reorders and retitles several examples in the 'Search template'
chapter. This is primarily to make examples for storing, deleting, and
using search templates more prominent.
* Change <templatename> to <templateid>
Several files in the REST APIs nav section are included using
:leveloffset: tags. This increments headings (h2 -> h3, h3 -> h4, etc.)
in those files and removes the :leveloffset: tags.
Other supporting changes:
* Alphabetizes top-level REST API nav items.
* Change 'indices APIs' heading to 'index APIs.'
* Changes 'Snapshot lifecycle management' heading to sentence case.
Moves the following API sections under the REST APIs navigations:
- API Conventions
- Document APIs
- Search APIs
- Index APIs (previously named Indices APIs)
- cat APIs
- Cluster APIs
Other supporting changes:
- Removes the previous index APIs page under REST APIs. Adds a redirect for the removed page.
- Removes several [partintro] macros so the docs build correctly.
- Changes anchors for pages that become sections of a parent page.
- Adds several redirects for existing pages that become sections of a parent page.
This commit re-applies changes from #44238. Changes from that PR were reverted due to broken links in several repos. This commit adds redirects for those broken links.
This commit removes the nested_path and nested_filter options deprecated in 6x.
This change also checks that the sort field has a [nested] option if it is under a nested
object and throws an exception if it's not the case.
Closes#27098
the geo-bounding-box and phrase-suggest docs were susceptible to
failing due to other indices in the cluster. This change restricts
the queries to the index that is set up for the test.
relates to #43271.
Adrien Grand
Mayya Sharipova
James Rodewig
Orhan Toy
Patrick Maynard
Christoph Büscher
Ian Danforth
Jim Ferenczi
István Zoltán Szabó
Alan Woodward
Takumasa Ochi
Nathan Howard
Jonathan Hult
Emmanuel DEMEY
Jesse Wright
Julie Tibshirani
John Murphy
Tal Levy