Skip to content

Conversation

@asimmahmood1
Copy link
Contributor

@asimmahmood1 asimmahmood1 commented Nov 19, 2025

Description

This is a built on top #19573, once that is merged will update this PR.

Related Issues

Resolves #19827
Part of #18882
Also #19384

Check List

  • Functionality includes testing.
  • [n/a] API changes companion pull request created, if applicable.
  • [TODO] Public documentation issue/PR created, if applicable.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.

jainankitk and others added 29 commits October 8, 2025 14:32
Signed-off-by: Ankit Jain <[email protected]>
Signed-off-by: Ankit Jain <[email protected]>
Signed-off-by: Ankit Jain <[email protected]>
Signed-off-by: Asim Mahmood <[email protected]>
Signed-off-by: Asim Mahmood <[email protected]>
…ent idea oh how to seperately filter rewrite path vs regular path.

Signed-off-by: Asim Mahmood <[email protected]>
Signed-off-by: Ankit Jain <[email protected]>
…gRangeCollector can call it out.

Signed-off-by: Asim Mahmood <[email protected]>
Signed-off-by: Asim Mahmood <[email protected]>
Signed-off-by: Asim Mahmood <[email protected]>
@github-actions github-actions bot added the v3.4.0 Issues and PRs related to version 3.4.0 label Nov 19, 2025
@asimmahmood1 asimmahmood1 self-assigned this Nov 19, 2025
@github-actions
Copy link
Contributor

❌ Gradle check result for 27f8248: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

@asimmahmood1
Copy link
Contributor Author

Functionally correct but not showing improvement

diff 17_big5_auto_date_filter_baseline.json 17_big5_auto_date_filter_candidate.json
2c2
<   "took": 114,
---
>   "took": 214,

Query

curl -X POST "http://localhost:9200/big5/_search"     -H "Content-Type: application/json"     -d '{
      "size": 0,
      "query": {
        "bool": {
          "must": [
            {
              "term": {
                "process.name": "systemd"
              }
            }
          ]
        }
      },
      "aggs": {
        "by_hour": {
          "auto_date_histogram": {
            "field": "@timestamp",
            "buckets": 3
          }
        }
      }
    }'

Result

{
  "took": 127,
  "timed_out": false,
  "terminated_early": true,
  "_shards": {
    "total": 1,
    "successful": 1,
    "skipped": 0,
    "failed": 0
  },
  "hits": {
    "total": {
      "value": 10000,
      "relation": "gte"
    },
    "max_score": null,
    "hits": []
  },
  "aggregations": {
    "by_hour": {
      "buckets": [
        {
          "key_as_string": "2023-01-01T00:00:00.000Z",
          "key": 1672531200000,
          "doc_count": 2488712
        },
        {
          "key_as_string": "2023-01-08T00:00:00.000Z",
          "key": 1673136000000,
          "doc_count": 824998
        }
      ],
      "interval": "7d"
    }
  }
}

@asimmahmood1
Copy link
Contributor Author

@jainankitk
Copy link
Contributor

@asimmahmood1 - Have you verified that the regression is only with AutoDateHistogramAggregator and not when using DateHistogramAggregator for same query?

…by filter rewrite

* TODO:
  * add more test cases, currently does not assert that skiplist was used
  * add auto date benchmark to http_logs
  * verify existing benchmarks have no issue

Signed-off-by: Asim Mahmood <[email protected]>
@asimmahmood1
Copy link
Contributor Author

asimmahmood1 commented Nov 24, 2025

So I figured out why performance was not up to par. Short answer is when auto date moves onto large intervals, we need to track of not just preparedRounding but also bucketOrds. If preparedRounding has changed since last time, we need to restart the skiplist logic. Otherwise, we'll collect too many docs, and although the end result doesn't change (i.e. unit test passes), performance is too low.

Auto date histo has two modes: FromSingle and FromMany. FromSingle is often used in top level aggregation, so It's similar to Date Histogram where parent is null. FromMany is used e.g in big5's range-auto-date-histo, which would normally handle interleaving owningBucketOrd. In the special case where filter rewrite logic is used, then we can safely assume that only 1 owningBucketOrd will be called per leaf collector, thus we can use skiplist histogram.

See validation below.

Note: will update this PR after #19573 is merged.

@asimmahmood1
Copy link
Contributor Author

range auto date: from 1335 to 139 (89%)

11_range_auto_date_histo.sh
#!/bin/bash

curl -XGET 'http://localhost:9200/big5/_search' \
-H 'Content-Type: application/json' \
-d '{
  "size": 0, "profile": false,
  "aggs": {
    "tmax": {
      "range": {
        "field": "metrics.size",
        "ranges": [
          {
            "to": -10
          },
          {
            "from": -10,
            "to": 10
          },
          {
            "from": 10,
            "to": 100
          },
          {
            "from": 100,
            "to": 1000
          },
          {
            "from": 1000,
            "to": 2000
          },
          {
            "from": 2000
          }
        ]
      },
      "aggs": {
        "date": {
          "auto_date_histogram": {
            "field": "@timestamp",
            "buckets": 20
          }
        }
      }
    }
  }
}'

diff 11_range_auto_date_histo_candidate.json 11_range_auto_date_histo_baseline.json
2c2
<   "took": 139,
---
>   "took": 1335,

@asimmahmood1
Copy link
Contributor Author

range-auto-date-with-metrics (22% lower)

this is similar to the date-with-metrics, since the time is bounded by tavg stat

[ec2-user@ip-172-31-61-197 ~]$ diff 11_range_auto_date_histo_with_metrics_candidate.json 11_range_auto_date_histo_with_metrics_baseline.json
2c2
<   "took": 2920,
---
>   "took": 3781,

Signed-off-by: Asim Mahmood <[email protected]>
@github-actions
Copy link
Contributor

❌ Gradle check result for 4d7f22d: FAILURE

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

@github-project-automation github-project-automation bot moved this from Todo to Done in Performance Roadmap Nov 24, 2025
@asimmahmood1 asimmahmood1 reopened this Nov 24, 2025
@github-project-automation github-project-automation bot moved this from Done to In Progress in Performance Roadmap Nov 24, 2025
@asimmahmood1
Copy link
Contributor Author

{"run-benchmark-test": "id_3"}

@asimmahmood1
Copy link
Contributor Author

{"run-benchmark-test": "id_11"}

@github-actions
Copy link
Contributor

❌ Gradle check result for 4d7f22d:

Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change?

@github-actions
Copy link
Contributor

The Jenkins job url is https://build.ci.opensearch.org/job/benchmark-pull-request/5174/ . Final results will be published once the job is completed.

@opensearch-ci-bot
Copy link
Collaborator

Benchmark Results

Benchmark Results for Job: https://build.ci.opensearch.org/job/benchmark-pull-request/5174/

Metric Task Value Unit
Cumulative indexing time of primary shards 0 min
Min cumulative indexing time across primary shards 0 min
Median cumulative indexing time across primary shards 0 min
Max cumulative indexing time across primary shards 0 min
Cumulative indexing throttle time of primary shards 0 min
Min cumulative indexing throttle time across primary shards 0 min
Median cumulative indexing throttle time across primary shards 0 min
Max cumulative indexing throttle time across primary shards 0 min
Cumulative merge time of primary shards 0 min
Cumulative merge count of primary shards 0
Min cumulative merge time across primary shards 0 min
Median cumulative merge time across primary shards 0 min
Max cumulative merge time across primary shards 0 min
Cumulative merge throttle time of primary shards 0 min
Min cumulative merge throttle time across primary shards 0 min
Median cumulative merge throttle time across primary shards 0 min
Max cumulative merge throttle time across primary shards 0 min
Cumulative refresh time of primary shards 0 min
Cumulative refresh count of primary shards 31
Min cumulative refresh time across primary shards 0 min
Median cumulative refresh time across primary shards 0 min
Max cumulative refresh time across primary shards 0 min
Cumulative flush time of primary shards 0 min
Cumulative flush count of primary shards 8
Min cumulative flush time across primary shards 0 min
Median cumulative flush time across primary shards 0 min
Max cumulative flush time across primary shards 0 min
Total Young Gen GC time 2.18 s
Total Young Gen GC count 71
Total Old Gen GC time 0 s
Total Old Gen GC count 0
Store size 15.3221 GB
Translog size 4.09782e-07 GB
Heap used for segments 0 MB
Heap used for doc values 0 MB
Heap used for terms 0 MB
Heap used for norms 0 MB
Heap used for points 0 MB
Heap used for stored fields 0 MB
Segment count 73
100th percentile latency wait-for-snapshot-recovery 300001 ms
100th percentile service time wait-for-snapshot-recovery 300001 ms
error rate wait-for-snapshot-recovery 100 %
Min Throughput match-all 8 ops/s
Mean Throughput match-all 8 ops/s
Median Throughput match-all 8 ops/s
Max Throughput match-all 8 ops/s
50th percentile latency match-all 4.28224 ms
90th percentile latency match-all 4.91128 ms
99th percentile latency match-all 5.84384 ms
100th percentile latency match-all 5.86495 ms
50th percentile service time match-all 3.42397 ms
90th percentile service time match-all 3.82077 ms
99th percentile service time match-all 4.51325 ms
100th percentile service time match-all 4.65042 ms
error rate match-all 0 %
Min Throughput term 49.9 ops/s
Mean Throughput term 49.9 ops/s
Median Throughput term 49.9 ops/s
Max Throughput term 49.91 ops/s
50th percentile latency term 3.64265 ms
90th percentile latency term 4.15728 ms
99th percentile latency term 9.31213 ms
100th percentile latency term 14.2462 ms
50th percentile service time term 2.90563 ms
90th percentile service time term 3.16569 ms
99th percentile service time term 8.3056 ms
100th percentile service time term 13.2171 ms
error rate term 0 %
Min Throughput range 1 ops/s
Mean Throughput range 1.01 ops/s
Median Throughput range 1.01 ops/s
Max Throughput range 1.01 ops/s
50th percentile latency range 6.06911 ms
90th percentile latency range 6.68034 ms
99th percentile latency range 7.30484 ms
100th percentile latency range 7.37473 ms
50th percentile service time range 4.3927 ms
90th percentile service time range 4.67829 ms
99th percentile service time range 5.61881 ms
100th percentile service time range 5.64028 ms
error rate range 0 %
Min Throughput 200s-in-range 32.93 ops/s
Mean Throughput 200s-in-range 32.93 ops/s
Median Throughput 200s-in-range 32.93 ops/s
Max Throughput 200s-in-range 32.94 ops/s
50th percentile latency 200s-in-range 5.10912 ms
90th percentile latency 200s-in-range 6.15178 ms
99th percentile latency 200s-in-range 7.16864 ms
100th percentile latency 200s-in-range 7.41385 ms
50th percentile service time 200s-in-range 3.93731 ms
90th percentile service time 200s-in-range 4.25686 ms
99th percentile service time 200s-in-range 5.041 ms
100th percentile service time 200s-in-range 5.14181 ms
error rate 200s-in-range 0 %
Min Throughput 400s-in-range 50.03 ops/s
Mean Throughput 400s-in-range 50.03 ops/s
Median Throughput 400s-in-range 50.03 ops/s
Max Throughput 400s-in-range 50.03 ops/s
50th percentile latency 400s-in-range 3.43545 ms
90th percentile latency 400s-in-range 3.88271 ms
99th percentile latency 400s-in-range 9.29701 ms
100th percentile latency 400s-in-range 14.4099 ms
50th percentile service time 400s-in-range 2.62956 ms
90th percentile service time 400s-in-range 2.79619 ms
99th percentile service time 400s-in-range 8.37331 ms
100th percentile service time 400s-in-range 13.362 ms
error rate 400s-in-range 0 %
Min Throughput hourly_agg 1.01 ops/s
Mean Throughput hourly_agg 1.01 ops/s
Median Throughput hourly_agg 1.01 ops/s
Max Throughput hourly_agg 1.02 ops/s
50th percentile latency hourly_agg 13.9039 ms
90th percentile latency hourly_agg 15.0932 ms
99th percentile latency hourly_agg 16.7868 ms
100th percentile latency hourly_agg 17.49 ms
50th percentile service time hourly_agg 12.0059 ms
90th percentile service time hourly_agg 13.004 ms
99th percentile service time hourly_agg 14.7653 ms
100th percentile service time hourly_agg 15.154 ms
error rate hourly_agg 0 %
Min Throughput hourly_agg_with_filter 1 ops/s
Mean Throughput hourly_agg_with_filter 1 ops/s
Median Throughput hourly_agg_with_filter 1 ops/s
Max Throughput hourly_agg_with_filter 1 ops/s
50th percentile latency hourly_agg_with_filter 83.5865 ms
90th percentile latency hourly_agg_with_filter 94.7305 ms
99th percentile latency hourly_agg_with_filter 140.956 ms
100th percentile latency hourly_agg_with_filter 183.002 ms
50th percentile service time hourly_agg_with_filter 81.8279 ms
90th percentile service time hourly_agg_with_filter 92.8934 ms
99th percentile service time hourly_agg_with_filter 139.147 ms
100th percentile service time hourly_agg_with_filter 181.211 ms
error rate hourly_agg_with_filter 0 %
Min Throughput hourly_agg_with_filter_and_metrics 0.24 ops/s
Mean Throughput hourly_agg_with_filter_and_metrics 0.24 ops/s
Median Throughput hourly_agg_with_filter_and_metrics 0.24 ops/s
Max Throughput hourly_agg_with_filter_and_metrics 0.24 ops/s
50th percentile latency hourly_agg_with_filter_and_metrics 323483 ms
90th percentile latency hourly_agg_with_filter_and_metrics 451170 ms
99th percentile latency hourly_agg_with_filter_and_metrics 479744 ms
100th percentile latency hourly_agg_with_filter_and_metrics 481334 ms
50th percentile service time hourly_agg_with_filter_and_metrics 4176.52 ms
90th percentile service time hourly_agg_with_filter_and_metrics 4281.44 ms
99th percentile service time hourly_agg_with_filter_and_metrics 4475.93 ms
100th percentile service time hourly_agg_with_filter_and_metrics 4558.98 ms
error rate hourly_agg_with_filter_and_metrics 0 %
Min Throughput multi_term_agg 0.22 ops/s
Mean Throughput multi_term_agg 0.22 ops/s
Median Throughput multi_term_agg 0.22 ops/s
Max Throughput multi_term_agg 0.23 ops/s
50th percentile latency multi_term_agg 347090 ms
90th percentile latency multi_term_agg 485132 ms
99th percentile latency multi_term_agg 516024 ms
100th percentile latency multi_term_agg 517783 ms
50th percentile service time multi_term_agg 4506.97 ms
90th percentile service time multi_term_agg 4626.37 ms
99th percentile service time multi_term_agg 4866.29 ms
100th percentile service time multi_term_agg 4986.95 ms
error rate multi_term_agg 0 %
Min Throughput scroll 25.04 pages/s
Mean Throughput scroll 25.07 pages/s
Median Throughput scroll 25.07 pages/s
Max Throughput scroll 25.13 pages/s
50th percentile latency scroll 207.343 ms
90th percentile latency scroll 210.843 ms
99th percentile latency scroll 264.671 ms
100th percentile latency scroll 291.279 ms
50th percentile service time scroll 205.387 ms
90th percentile service time scroll 208.758 ms
99th percentile service time scroll 262.554 ms
100th percentile service time scroll 289.84 ms
error rate scroll 0 %
Min Throughput desc_sort_size 1 ops/s
Mean Throughput desc_sort_size 1 ops/s
Median Throughput desc_sort_size 1 ops/s
Max Throughput desc_sort_size 1 ops/s
50th percentile latency desc_sort_size 7.20233 ms
90th percentile latency desc_sort_size 7.99693 ms
99th percentile latency desc_sort_size 8.81847 ms
100th percentile latency desc_sort_size 8.96338 ms
50th percentile service time desc_sort_size 5.40942 ms
90th percentile service time desc_sort_size 5.89622 ms
99th percentile service time desc_sort_size 6.66956 ms
100th percentile service time desc_sort_size 6.68186 ms
error rate desc_sort_size 0 %
Min Throughput asc_sort_size 1 ops/s
Mean Throughput asc_sort_size 1 ops/s
Median Throughput asc_sort_size 1 ops/s
Max Throughput asc_sort_size 1 ops/s
50th percentile latency asc_sort_size 8.26643 ms
90th percentile latency asc_sort_size 8.92356 ms
99th percentile latency asc_sort_size 9.54172 ms
100th percentile latency asc_sort_size 9.60887 ms
50th percentile service time asc_sort_size 6.35152 ms
90th percentile service time asc_sort_size 7.0678 ms
99th percentile service time asc_sort_size 7.44685 ms
100th percentile service time asc_sort_size 7.53512 ms
error rate asc_sort_size 0 %
Min Throughput desc_sort_timestamp 1 ops/s
Mean Throughput desc_sort_timestamp 1 ops/s
Median Throughput desc_sort_timestamp 1 ops/s
Max Throughput desc_sort_timestamp 1 ops/s
50th percentile latency desc_sort_timestamp 13.5088 ms
90th percentile latency desc_sort_timestamp 14.2314 ms
99th percentile latency desc_sort_timestamp 15.988 ms
100th percentile latency desc_sort_timestamp 16.0292 ms
50th percentile service time desc_sort_timestamp 11.8524 ms
90th percentile service time desc_sort_timestamp 12.2587 ms
99th percentile service time desc_sort_timestamp 14.5756 ms
100th percentile service time desc_sort_timestamp 14.6071 ms
error rate desc_sort_timestamp 0 %
Min Throughput asc_sort_timestamp 1 ops/s
Mean Throughput asc_sort_timestamp 1 ops/s
Median Throughput asc_sort_timestamp 1 ops/s
Max Throughput asc_sort_timestamp 1 ops/s
50th percentile latency asc_sort_timestamp 8.08182 ms
90th percentile latency asc_sort_timestamp 8.76246 ms
99th percentile latency asc_sort_timestamp 9.50709 ms
100th percentile latency asc_sort_timestamp 10.022 ms
50th percentile service time asc_sort_timestamp 6.20787 ms
90th percentile service time asc_sort_timestamp 6.69686 ms
99th percentile service time asc_sort_timestamp 7.68305 ms
100th percentile service time asc_sort_timestamp 8.12289 ms
error rate asc_sort_timestamp 0 %
Min Throughput desc_sort_with_after_timestamp 1.01 ops/s
Mean Throughput desc_sort_with_after_timestamp 1.02 ops/s
Median Throughput desc_sort_with_after_timestamp 1.02 ops/s
Max Throughput desc_sort_with_after_timestamp 1.1 ops/s
50th percentile latency desc_sort_with_after_timestamp 6.27938 ms
90th percentile latency desc_sort_with_after_timestamp 6.80572 ms
99th percentile latency desc_sort_with_after_timestamp 7.46574 ms
100th percentile latency desc_sort_with_after_timestamp 7.57326 ms
50th percentile service time desc_sort_with_after_timestamp 4.42068 ms
90th percentile service time desc_sort_with_after_timestamp 4.76673 ms
99th percentile service time desc_sort_with_after_timestamp 5.54311 ms
100th percentile service time desc_sort_with_after_timestamp 5.58049 ms
error rate desc_sort_with_after_timestamp 0 %
Min Throughput asc_sort_with_after_timestamp 1.01 ops/s
Mean Throughput asc_sort_with_after_timestamp 1.02 ops/s
Median Throughput asc_sort_with_after_timestamp 1.02 ops/s
Max Throughput asc_sort_with_after_timestamp 1.1 ops/s
50th percentile latency asc_sort_with_after_timestamp 5.39147 ms
90th percentile latency asc_sort_with_after_timestamp 5.85555 ms
99th percentile latency asc_sort_with_after_timestamp 6.20304 ms
100th percentile latency asc_sort_with_after_timestamp 6.30199 ms
50th percentile service time asc_sort_with_after_timestamp 3.60652 ms
90th percentile service time asc_sort_with_after_timestamp 3.74439 ms
99th percentile service time asc_sort_with_after_timestamp 3.92286 ms
100th percentile service time asc_sort_with_after_timestamp 4.0068 ms
error rate asc_sort_with_after_timestamp 0 %
Min Throughput range_size 2.01 ops/s
Mean Throughput range_size 2.01 ops/s
Median Throughput range_size 2.01 ops/s
Max Throughput range_size 2.02 ops/s
50th percentile latency range_size 8.27167 ms
90th percentile latency range_size 8.83461 ms
99th percentile latency range_size 10.0267 ms
100th percentile latency range_size 10.1073 ms
50th percentile service time range_size 6.9538 ms
90th percentile service time range_size 7.30531 ms
99th percentile service time range_size 8.47684 ms
100th percentile service time range_size 8.55189 ms
error rate range_size 0 %
Min Throughput range_with_asc_sort 2.01 ops/s
Mean Throughput range_with_asc_sort 2.01 ops/s
Median Throughput range_with_asc_sort 2.01 ops/s
Max Throughput range_with_asc_sort 2.02 ops/s
50th percentile latency range_with_asc_sort 18.8966 ms
90th percentile latency range_with_asc_sort 20.8522 ms
99th percentile latency range_with_asc_sort 22.2688 ms
100th percentile latency range_with_asc_sort 22.4174 ms
50th percentile service time range_with_asc_sort 17.4148 ms
90th percentile service time range_with_asc_sort 19.218 ms
99th percentile service time range_with_asc_sort 20.434 ms
100th percentile service time range_with_asc_sort 20.5176 ms
error rate range_with_asc_sort 0 %
Min Throughput range_with_desc_sort 2.01 ops/s
Mean Throughput range_with_desc_sort 2.01 ops/s
Median Throughput range_with_desc_sort 2.01 ops/s
Max Throughput range_with_desc_sort 2.02 ops/s
50th percentile latency range_with_desc_sort 20.7926 ms
90th percentile latency range_with_desc_sort 24.532 ms
99th percentile latency range_with_desc_sort 33.4369 ms
100th percentile latency range_with_desc_sort 41.1489 ms
50th percentile service time range_with_desc_sort 18.6428 ms
90th percentile service time range_with_desc_sort 22.7379 ms
99th percentile service time range_with_desc_sort 31.1639 ms
100th percentile service time range_with_desc_sort 38.9406 ms
error rate range_with_desc_sort 0 %

@opensearch-ci-bot
Copy link
Collaborator

Benchmark Baseline Comparison Results

Benchmark Results for Job: https://build.ci.opensearch.org/job/benchmark-compare/210/

Metric Task Baseline Contender Diff Unit
Cumulative indexing time of primary shards 0 0 0 min
Min cumulative indexing time across primary shard 0 0 0 min
Median cumulative indexing time across primary shard 0 0 0 min
Max cumulative indexing time across primary shard 0 0 0 min
Cumulative indexing throttle time of primary shards 0 0 0 min
Min cumulative indexing throttle time across primary shard 0 0 0 min
Median cumulative indexing throttle time across primary shard 0 0 0 min
Max cumulative indexing throttle time across primary shard 0 0 0 min
Cumulative merge time of primary shards 0 0 0 min
Cumulative merge count of primary shards 0 0 0
Min cumulative merge time across primary shard 0 0 0 min
Median cumulative merge time across primary shard 0 0 0 min
Max cumulative merge time across primary shard 0 0 0 min
Cumulative merge throttle time of primary shards 0 0 0 min
Min cumulative merge throttle time across primary shard 0 0 0 min
Median cumulative merge throttle time across primary shard 0 0 0 min
Max cumulative merge throttle time across primary shard 0 0 0 min
Cumulative refresh time of primary shards 0 0 0 min
Cumulative refresh count of primary shards 31 31 0
Min cumulative refresh time across primary shard 0 0 0 min
Median cumulative refresh time across primary shard 0 0 0 min
Max cumulative refresh time across primary shard 0 0 0 min
Cumulative flush time of primary shards 0 0 0 min
Cumulative flush count of primary shards 8 8 0
Min cumulative flush time across primary shard 0 0 0 min
Median cumulative flush time across primary shard 0 0 0 min
Max cumulative flush time across primary shard 0 0 0 min
Total Young Gen GC time 2.16 2.18 0.02 s
Total Young Gen GC count 71 71 0
Total Old Gen GC time 0 0 0 s
Total Old Gen GC count 0 0 0
Store size 15.3221 15.3221 0 GB
Translog size 4.09782e-07 4.09782e-07 0 GB
Heap used for segments 0 0 0 MB
Heap used for doc values 0 0 0 MB
Heap used for terms 0 0 0 MB
Heap used for norms 0 0 0 MB
Heap used for points 0 0 0 MB
Heap used for stored fields 0 0 0 MB
Segment count 73 73 0
100th percentile latency wait-for-snapshot-recovery 300002 300001 -0.46875 ms
100th percentile service time wait-for-snapshot-recovery 300002 300001 -0.46875 ms
error rate wait-for-snapshot-recovery 100 100 0 %
Min Throughput match-all 8.00004 7.99863 -0.00142 ops/s
Mean Throughput match-all 8.0001 7.99876 -0.00134 ops/s
Median Throughput match-all 8.00011 7.99878 -0.00133 ops/s
Max Throughput match-all 8.00013 7.99891 -0.00122 ops/s
50th percentile latency match-all 4.08353 4.28224 0.1987 ms
90th percentile latency match-all 4.69237 4.91128 0.21891 ms
99th percentile latency match-all 5.07276 5.84384 0.77108 ms
100th percentile latency match-all 5.15329 5.86495 0.71166 ms
50th percentile service time match-all 3.09253 3.42397 0.33145 ms
90th percentile service time match-all 3.56093 3.82077 0.25984 ms
99th percentile service time match-all 4.31898 4.51325 0.19427 ms
100th percentile service time match-all 4.38225 4.65042 0.26817 ms
error rate match-all 0 0 0 %
Min Throughput term 49.8653 49.898 0.03277 ops/s
Mean Throughput term 49.8705 49.9015 0.03104 ops/s
Median Throughput term 49.8705 49.9015 0.03104 ops/s
Max Throughput term 49.8757 49.905 0.02931 ops/s
50th percentile latency term 3.45205 3.64265 0.1906 ms
90th percentile latency term 3.89325 4.15728 0.26403 ms
99th percentile latency term 8.97703 9.31213 0.3351 ms
100th percentile latency term 13.9687 14.2462 0.27751 ms
50th percentile service time term 2.65867 2.90563 0.24697 ms
90th percentile service time term 2.84279 3.16569 0.32289 ms
99th percentile service time term 3.18447 8.3056 5.12113 ms
100th percentile service time term 3.20051 13.2171 10.0166 ms
error rate term 0 0 0 %
Min Throughput range 1.00478 1.00465 -0.00013 ops/s
Mean Throughput range 1.00662 1.00644 -0.00018 ops/s
Median Throughput range 1.00636 1.00619 -0.00018 ops/s
Max Throughput range 1.00951 1.00925 -0.00026 ops/s
50th percentile latency range 6.32679 6.06911 -0.25767 ms
90th percentile latency range 6.74759 6.68034 -0.06725 ms
99th percentile latency range 14.9304 7.30484 -7.62552 ms
100th percentile latency range 22.2679 7.37473 -14.8932 ms
50th percentile service time range 4.40336 4.3927 -0.01066 ms
90th percentile service time range 4.713 4.67829 -0.03471 ms
99th percentile service time range 13.1272 5.61881 -7.50838 ms
100th percentile service time range 20.2359 5.64028 -14.5956 ms
error rate range 0 0 0 %
Min Throughput 200s-in-range 32.9022 32.9316 0.02934 ops/s
Mean Throughput 200s-in-range 32.9078 32.9337 0.02591 ops/s
Median Throughput 200s-in-range 32.9084 32.9323 0.0239 ops/s
Max Throughput 200s-in-range 32.9129 32.9374 0.02448 ops/s
50th percentile latency 200s-in-range 4.86638 5.10912 0.24274 ms
90th percentile latency 200s-in-range 5.66428 6.15178 0.4875 ms
99th percentile latency 200s-in-range 6.22094 7.16864 0.9477 ms
100th percentile latency 200s-in-range 6.60392 7.41385 0.80994 ms
50th percentile service time 200s-in-range 3.52999 3.93731 0.40733 ms
90th percentile service time 200s-in-range 3.69637 4.25686 0.5605 ms
99th percentile service time 200s-in-range 4.96872 5.041 0.07228 ms
100th percentile service time 200s-in-range 5.87261 5.14181 -0.7308 ms
error rate 200s-in-range 0 0 0 %
Min Throughput 400s-in-range 50.0106 50.033 0.02244 ops/s
Mean Throughput 400s-in-range 50.0119 50.034 0.02205 ops/s
Median Throughput 400s-in-range 50.0119 50.034 0.02205 ops/s
Max Throughput 400s-in-range 50.0132 50.0349 0.02166 ops/s
50th percentile latency 400s-in-range 3.69972 3.43545 -0.26427 ms
90th percentile latency 400s-in-range 4.12578 3.88271 -0.24306 ms
99th percentile latency 400s-in-range 9.65137 9.29701 -0.35436 ms
100th percentile latency 400s-in-range 14.8397 14.4099 -0.42975 ms
50th percentile service time 400s-in-range 2.91395 2.62956 -0.28439 ms
90th percentile service time 400s-in-range 3.02216 2.79619 -0.22597 ms
99th percentile service time 400s-in-range 8.69587 8.37331 -0.32257 ms
100th percentile service time 400s-in-range 13.9622 13.362 -0.60017 ms
error rate 400s-in-range 0 0 0 %
Min Throughput hourly_agg 1.00566 1.00571 5e-05 ops/s
Mean Throughput hourly_agg 1.00932 1.0094 8e-05 ops/s
Median Throughput hourly_agg 1.00848 1.00855 7e-05 ops/s
Max Throughput hourly_agg 1.01684 1.01699 0.00016 ops/s
50th percentile latency hourly_agg 13.2876 13.9039 0.61626 ms
90th percentile latency hourly_agg 14.3772 15.0932 0.71602 ms
99th percentile latency hourly_agg 16.4744 16.7868 0.31244 ms
100th percentile latency hourly_agg 16.7372 17.49 0.75281 ms
50th percentile service time hourly_agg 11.4508 12.0059 0.55507 ms
90th percentile service time hourly_agg 12.4074 13.004 0.59659 ms
99th percentile service time hourly_agg 14.3855 14.7653 0.37978 ms
100th percentile service time hourly_agg 14.8684 15.154 0.28559 ms
error rate hourly_agg 0 0 0 %
Min Throughput hourly_agg_with_filter 1.00298 1.00122 -0.00176 ops/s
Mean Throughput hourly_agg_with_filter 1.00488 1.002 -0.00288 ops/s
Median Throughput hourly_agg_with_filter 1.00445 1.00182 -0.00262 ops/s
Max Throughput hourly_agg_with_filter 1.00879 1.00361 -0.00518 ops/s
50th percentile latency hourly_agg_with_filter 81.6363 83.5865 1.95017 ms
90th percentile latency hourly_agg_with_filter 92.412 94.7305 2.31843 ms
99th percentile latency hourly_agg_with_filter 127.853 140.956 13.1029 ms
100th percentile latency hourly_agg_with_filter 160.239 183.002 22.7638 ms
50th percentile service time hourly_agg_with_filter 79.6116 81.8279 2.21627 ms
90th percentile service time hourly_agg_with_filter 90.3268 92.8934 2.56658 ms
99th percentile service time hourly_agg_with_filter 126 139.147 13.1461 ms
100th percentile service time hourly_agg_with_filter 158.385 181.211 22.8264 ms
error rate hourly_agg_with_filter 0 0 0 %
Min Throughput hourly_agg_with_filter_and_metrics 0.216111 0.235085 0.01897 ops/s
Mean Throughput hourly_agg_with_filter_and_metrics 0.216971 0.236491 0.01952 ops/s
Median Throughput hourly_agg_with_filter_and_metrics 0.216991 0.236566 0.01958 ops/s
Max Throughput hourly_agg_with_filter_and_metrics 0.217794 0.237297 0.0195 ops/s
50th percentile latency hourly_agg_with_filter_and_metrics 363245 323483 -39762.4 ms
90th percentile latency hourly_agg_with_filter_and_metrics 505752 451170 -54581.8 ms
99th percentile latency hourly_agg_with_filter_and_metrics 538132 479744 -58388.8 ms
100th percentile latency hourly_agg_with_filter_and_metrics 539921 481334 -58586.9 ms
50th percentile service time hourly_agg_with_filter_and_metrics 4576.64 4176.52 -400.121 ms
90th percentile service time hourly_agg_with_filter_and_metrics 4684.03 4281.44 -402.596 ms
99th percentile service time hourly_agg_with_filter_and_metrics 4785.3 4475.93 -309.373 ms
100th percentile service time hourly_agg_with_filter_and_metrics 4817.81 4558.98 -258.823 ms
error rate hourly_agg_with_filter_and_metrics 0 0 0 %
Min Throughput multi_term_agg 0.220417 0.224183 0.00377 ops/s
Mean Throughput multi_term_agg 0.222457 0.224716 0.00226 ops/s
Median Throughput multi_term_agg 0.222732 0.224588 0.00186 ops/s
Max Throughput multi_term_agg 0.223353 0.226172 0.00282 ops/s
50th percentile latency multi_term_agg 350647 347090 -3557.05 ms
90th percentile latency multi_term_agg 489341 485132 -4208.94 ms
99th percentile latency multi_term_agg 520187 516024 -4163.12 ms
100th percentile latency multi_term_agg 521879 517783 -4096.09 ms
50th percentile service time multi_term_agg 4483.31 4506.97 23.6614 ms
90th percentile service time multi_term_agg 4644.73 4626.37 -18.353 ms
99th percentile service time multi_term_agg 4690.78 4866.29 175.507 ms
100th percentile service time multi_term_agg 4711.89 4986.95 275.058 ms
error rate multi_term_agg 0 0 0 %
Min Throughput scroll 25.0498 25.0438 -0.00595 pages/s
Mean Throughput scroll 25.0819 25.0721 -0.00981 pages/s
Median Throughput scroll 25.0745 25.0656 -0.0089 pages/s
Max Throughput scroll 25.1485 25.1306 -0.0179 pages/s
50th percentile latency scroll 209.486 207.343 -2.14297 ms
90th percentile latency scroll 214.075 210.843 -3.23216 ms
99th percentile latency scroll 260.846 264.671 3.82584 ms
100th percentile latency scroll 283.995 291.279 7.28326 ms
50th percentile service time scroll 207.623 205.387 -2.23612 ms
90th percentile service time scroll 211.888 208.758 -3.12957 ms
99th percentile service time scroll 258.878 262.554 3.67609 ms
100th percentile service time scroll 281.69 289.84 8.14999 ms
error rate scroll 0 0 0 %
Min Throughput desc_sort_size 1.00319 1.0032 1e-05 ops/s
Mean Throughput desc_sort_size 1.00388 1.00389 1e-05 ops/s
Median Throughput desc_sort_size 1.00383 1.00384 1e-05 ops/s
Max Throughput desc_sort_size 1.00478 1.00479 1e-05 ops/s
50th percentile latency desc_sort_size 7.67193 7.20233 -0.4696 ms
90th percentile latency desc_sort_size 8.26067 7.99693 -0.26374 ms
99th percentile latency desc_sort_size 9.19009 8.81847 -0.37162 ms
100th percentile latency desc_sort_size 9.25557 8.96338 -0.29219 ms
50th percentile service time desc_sort_size 5.80691 5.40942 -0.39748 ms
90th percentile service time desc_sort_size 6.34789 5.89622 -0.45167 ms
99th percentile service time desc_sort_size 7.17157 6.66956 -0.502 ms
100th percentile service time desc_sort_size 7.36431 6.68186 -0.68245 ms
error rate desc_sort_size 0 0 0 %
Min Throughput asc_sort_size 1.0032 1.00323 3e-05 ops/s
Mean Throughput asc_sort_size 1.00389 1.00392 3e-05 ops/s
Median Throughput asc_sort_size 1.00384 1.00387 3e-05 ops/s
Max Throughput asc_sort_size 1.00479 1.00483 4e-05 ops/s
50th percentile latency asc_sort_size 8.39235 8.26643 -0.12592 ms
90th percentile latency asc_sort_size 9.23733 8.92356 -0.31377 ms
99th percentile latency asc_sort_size 9.98721 9.54172 -0.4455 ms
100th percentile latency asc_sort_size 9.99685 9.60887 -0.38799 ms
50th percentile service time asc_sort_size 6.65887 6.35152 -0.30736 ms
90th percentile service time asc_sort_size 7.37027 7.0678 -0.30247 ms
99th percentile service time asc_sort_size 8.08429 7.44685 -0.63744 ms
100th percentile service time asc_sort_size 8.30731 7.53512 -0.77219 ms
error rate asc_sort_size 0 0 0 %
Min Throughput desc_sort_timestamp 1.00316 1.00312 -4e-05 ops/s
Mean Throughput desc_sort_timestamp 1.00384 1.00379 -5e-05 ops/s
Median Throughput desc_sort_timestamp 1.00378 1.00374 -5e-05 ops/s
Max Throughput desc_sort_timestamp 1.00472 1.00466 -6e-05 ops/s
50th percentile latency desc_sort_timestamp 13.7684 13.5088 -0.25965 ms
90th percentile latency desc_sort_timestamp 14.6248 14.2314 -0.39336 ms
99th percentile latency desc_sort_timestamp 16.3723 15.988 -0.3843 ms
100th percentile latency desc_sort_timestamp 16.813 16.0292 -0.78384 ms
50th percentile service time desc_sort_timestamp 12.0042 11.8524 -0.15178 ms
90th percentile service time desc_sort_timestamp 12.5057 12.2587 -0.24695 ms
99th percentile service time desc_sort_timestamp 14.3426 14.5756 0.233 ms
100th percentile service time desc_sort_timestamp 14.6647 14.6071 -0.05765 ms
error rate desc_sort_timestamp 0 0 0 %
Min Throughput asc_sort_timestamp 1.00327 1.00328 0 ops/s
Mean Throughput asc_sort_timestamp 1.00398 1.00398 0 ops/s
Median Throughput asc_sort_timestamp 1.00392 1.00393 0 ops/s
Max Throughput asc_sort_timestamp 1.00489 1.0049 1e-05 ops/s
50th percentile latency asc_sort_timestamp 7.98311 8.08182 0.09871 ms
90th percentile latency asc_sort_timestamp 8.56813 8.76246 0.19433 ms
99th percentile latency asc_sort_timestamp 9.33831 9.50709 0.16878 ms
100th percentile latency asc_sort_timestamp 9.5369 10.022 0.48514 ms
50th percentile service time asc_sort_timestamp 5.97092 6.20787 0.23695 ms
90th percentile service time asc_sort_timestamp 6.56714 6.69686 0.12972 ms
99th percentile service time asc_sort_timestamp 7.26511 7.68305 0.41794 ms
100th percentile service time asc_sort_timestamp 7.36109 8.12289 0.7618 ms
error rate asc_sort_timestamp 0 0 0 %
Min Throughput desc_sort_with_after_timestamp 1.00902 1.00899 -3e-05 ops/s
Mean Throughput desc_sort_with_after_timestamp 1.02402 1.02394 -8e-05 ops/s
Median Throughput desc_sort_with_after_timestamp 1.01652 1.01647 -5e-05 ops/s
Max Throughput desc_sort_with_after_timestamp 1.09819 1.09782 -0.00036 ops/s
50th percentile latency desc_sort_with_after_timestamp 5.98317 6.27938 0.29621 ms
90th percentile latency desc_sort_with_after_timestamp 6.5393 6.80572 0.26642 ms
99th percentile latency desc_sort_with_after_timestamp 6.8667 7.46574 0.59904 ms
100th percentile latency desc_sort_with_after_timestamp 6.88107 7.57326 0.69219 ms
50th percentile service time desc_sort_with_after_timestamp 4.20681 4.42068 0.21387 ms
90th percentile service time desc_sort_with_after_timestamp 4.5541 4.76673 0.21263 ms
99th percentile service time desc_sort_with_after_timestamp 5.08222 5.54311 0.46089 ms
100th percentile service time desc_sort_with_after_timestamp 5.23009 5.58049 0.35039 ms
error rate desc_sort_with_after_timestamp 0 0 0 %
Min Throughput asc_sort_with_after_timestamp 1.00906 1.00906 -0 ops/s
Mean Throughput asc_sort_with_after_timestamp 1.02412 1.02411 -0 ops/s
Median Throughput asc_sort_with_after_timestamp 1.01659 1.01659 -0 ops/s
Max Throughput asc_sort_with_after_timestamp 1.09864 1.09855 -9e-05 ops/s
50th percentile latency asc_sort_with_after_timestamp 5.53987 5.39147 -0.1484 ms
90th percentile latency asc_sort_with_after_timestamp 5.91514 5.85555 -0.05959 ms
99th percentile latency asc_sort_with_after_timestamp 6.20781 6.20304 -0.00477 ms
100th percentile latency asc_sort_with_after_timestamp 6.20875 6.30199 0.09324 ms
50th percentile service time asc_sort_with_after_timestamp 3.65936 3.60652 -0.05284 ms
90th percentile service time asc_sort_with_after_timestamp 3.81855 3.74439 -0.07415 ms
99th percentile service time asc_sort_with_after_timestamp 3.94172 3.92286 -0.01886 ms
100th percentile service time asc_sort_with_after_timestamp 3.94465 4.0068 0.06214 ms
error rate asc_sort_with_after_timestamp 0 0 0 %
Min Throughput range_size 2.00953 2.00955 1e-05 ops/s
Mean Throughput range_size 2.01319 2.0132 1e-05 ops/s
Median Throughput range_size 2.01268 2.01269 1e-05 ops/s
Max Throughput range_size 2.01888 2.0189 3e-05 ops/s
50th percentile latency range_size 8.52367 8.27167 -0.25199 ms
90th percentile latency range_size 9.10341 8.83461 -0.2688 ms
99th percentile latency range_size 9.8454 10.0267 0.18131 ms
100th percentile latency range_size 10.0394 10.1073 0.06786 ms
50th percentile service time range_size 7.26068 6.9538 -0.30688 ms
90th percentile service time range_size 7.49863 7.30531 -0.19332 ms
99th percentile service time range_size 8.3443 8.47684 0.13253 ms
100th percentile service time range_size 8.40524 8.55189 0.14665 ms
error rate range_size 0 0 0 %
Min Throughput range_with_asc_sort 2.00853 2.00832 -0.00022 ops/s
Mean Throughput range_with_asc_sort 2.01181 2.01152 -0.00029 ops/s
Median Throughput range_with_asc_sort 2.01135 2.01108 -0.00027 ops/s
Max Throughput range_with_asc_sort 2.01693 2.0165 -0.00043 ops/s
50th percentile latency range_with_asc_sort 19.3013 18.8966 -0.40474 ms
90th percentile latency range_with_asc_sort 21.44 20.8522 -0.5878 ms
99th percentile latency range_with_asc_sort 22.3658 22.2688 -0.097 ms
100th percentile latency range_with_asc_sort 22.408 22.4174 0.00942 ms
50th percentile service time range_with_asc_sort 17.5966 17.4148 -0.18179 ms
90th percentile service time range_with_asc_sort 19.992 19.218 -0.77396 ms
99th percentile service time range_with_asc_sort 20.5763 20.434 -0.14236 ms
100th percentile service time range_with_asc_sort 20.6325 20.5176 -0.11491 ms
error rate range_with_asc_sort 0 0 0 %
Min Throughput range_with_desc_sort 2.0093 2.00933 3e-05 ops/s
Mean Throughput range_with_desc_sort 2.01286 2.0129 4e-05 ops/s
Median Throughput range_with_desc_sort 2.01237 2.01242 6e-05 ops/s
Max Throughput range_with_desc_sort 2.01843 2.01852 9e-05 ops/s
50th percentile latency range_with_desc_sort 20.9821 20.7926 -0.18952 ms
90th percentile latency range_with_desc_sort 24.4813 24.532 0.0507 ms
99th percentile latency range_with_desc_sort 25.4849 33.4369 7.95205 ms
100th percentile latency range_with_desc_sort 25.504 41.1489 15.6449 ms
50th percentile service time range_with_desc_sort 18.6587 18.6428 -0.01594 ms
90th percentile service time range_with_desc_sort 22.2279 22.7379 0.51001 ms
99th percentile service time range_with_desc_sort 23.3495 31.1639 7.81438 ms
100th percentile service time range_with_desc_sort 23.4663 38.9406 15.4744 ms
error rate range_with_desc_sort 0 0 0 %

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Search:Aggregations Search:Performance v3.4.0 Issues and PRs related to version 3.4.0

Projects

Status: In Progress

Development

Successfully merging this pull request may close these issues.

Add skip_list logic to auto date histogram, confirm with big5's range-auto-date-history-with-metrcis, range-auto-date-history

3 participants