产品动态
产品公告
安全公告
Kona JDK
Kona JDK 介绍
Kona JDK 是什么？
Tencent Kona JDK，是腾讯基于社区 Open JDK 定制开发的 JDK 版本，广泛服务于腾讯内部业务和腾讯云上客户，经过了内部大数据和AI等复杂业务场景的验证，为腾讯 JAVA 生态提供专业持续的保障，具有稳定性高、安全性高、性能好等特点。
Kona JDK 优势
高性能和低成本：在腾讯大数据计算场景数万台服务器生产集群验证下，和Open JDK 8相比，Kona JDK 8吞吐提升8%，CPU 和内存使用率均降低10%左右；
开箱即用的 Vector API 支持：解决了向量指令适配导致的 JVM crash 等问题，业界率先落地，稳定支持广告训练场景；
多种 GC 优化：G1 GC 内存 overhead 和并行 Full GC 算法优化，同时针对强实时在线服务需求，推出生产级别的 ZGC；
KonaFiber 协程：已经在IEG天美游戏业务合作落地，目前根据benchmark测试，协程创建/切换/调度等性能大幅超过社区Loom；
支持 ARM 国产化 CPU；
Kona JDK 设置
1. 操作页面：集群详情页-高级配置页面。
2. 重启：切换JDK后，集群会重启，重启过程中不可进行变配升级等操作。
数据对比
本文提供4 核16G 3个节点的腾讯云 Elasticsearch Service 集群 切换 腾讯自研 Kona JDK + G1-GC 的集群性能指标。
说明：
数据为官方提供的 geonames，包含11396503条地理位置信息，总大小接近3GB，包含了 text、long、geo 等类型数据，覆盖行列存。
对比4核16GB SSD 200G 3节点的腾讯云 ES在 Kona JDK（11.0.9.1-ga+1） + G1-GC 和 Oracle JDK （1.8.0_181-b13）+CMS-GC 的集群性能指标，Kona JDK + G1-GC 各方面性能均有一定优势。主要得益于腾讯云自研 JDK 以及 GC 参数调优等方面的优化。
说明 | Metric | Task | 腾讯云ES(Open JDK+CMS-GC) | 腾讯云ES(Kona JDK+G1-GC) | 差异值（Kona JDK-Open JDK） | Unit | 优劣 |
写入总耗时 | Cumulative indexing time of primary shards | | 17.7745 | 17.41703333 | -0.35747 | min | 优 |
GC 总次数、耗时统计 | Total Young Gen GC time | | 76.597 | 4.988 | -71.609 | s | 优 |
Total Young Gen GC count | | 4129 | 981 | -3148 | | | |
Total Old Gen GC time | | 0.175 | 0 | -0.175 | s | | |
Total Old Gen GC count | | 2 | 0 | -2 | | | |
存储大小 | Store size | | 2.885286688 | 2.87756444 | -0.00772 | GB | - |
Translog size | | 3.59E-07 | 3.59E-07 | 0 | GB | | |
堆内存使用量 | Heap used for segments | | 0.045909882 | 0.045909882 | 0 | MB | |
Heap used for doc values | | 0.000507355 | 0.000507355 | 0 | MB | | |
Heap used for terms | | 0.037261963 | 0.037261963 | 0 | MB | | |
Heap used for norms | | 0.003967285 | 0.003967285 | 0 | MB | | |
Heap used for points | | 0 | 0 | 0 | MB | | |
Heap used for stored fields | | 0.004173279 | 0.004173279 | 0 | MB | | |
segment 总数量 | Segment count | | 7 | 7 | 0 | | |
写入吞吐、耗时统计 | Min Throughput | index-append | 82341.03288 | 83678.1806 | 1337.14773 | docs/s | 优 |
Mean Throughput | index-append | 85463.64521 | 87617.83008 | 2154.18488 | docs/s | | |
Median Throughput | index-append | 85203.41999 | 87749.05385 | 2545.63387 | docs/s | | |
Max Throughput | index-append | 88282.10166 | 90448.56087 | 2166.45921 | docs/s | | |
50th percentile latency | index-append | 369.9228433 | 360.6725829 | -9.25026 | ms | | |
90th percentile latency | index-append | 582.2157889 | 521.3938987 | -60.82189 | ms | | |
99th percentile latency | index-append | 2566.001355 | 3331.056718 | 765.05536 | ms | | |
99.9th percentile latency | index-append | 3249.023346 | 4277.054507 | 1028.03116 | ms | | |
100th percentile latency | index-append | 3677.799375 | 5966.865065 | 2289.06569 | ms | | |
50th percentile service time | index-append | 369.9228433 | 360.6725829 | -9.25026 | ms | | |
90th percentile service time | index-append | 582.2157889 | 521.3938987 | -60.82189 | ms | | |
99th percentile service time | index-append | 2566.001355 | 3331.056718 | 765.05536 | ms | | |
99.9th percentile service time | index-append | 3249.023346 | 4277.054507 | 1028.03116 | ms | | |
100th percentile service time | index-append | 3677.799375 | 5966.865065 | 2289.06569 | ms | | |
error rate | index-append | 0 | 0 | 0 | % | | |
index 指标统计 | Min Throughput | index-stats | 90.00917629 | 90.01515386 | 0.00598 | ops/s | |
Mean Throughput | index-stats | 90.01643728 | 90.02981242 | 0.01338 | ops/s | | |
Median Throughput | index-stats | 90.01545276 | 90.03117871 | 0.01573 | ops/s | | |
Max Throughput | index-stats | 90.0358266 | 90.0417802 | 0.00595 | ops/s | | |
50th percentile latency | index-stats | 2.71807611 | 2.706557978 | -0.01152 | ms | | |
90th percentile latency | index-stats | 3.542617336 | 3.530823253 | -0.01179 | ms | | |
99th percentile latency | index-stats | 4.209796032 | 4.047978334 | -0.16182 | ms | | |
99.9th percentile latency | index-stats | 10.97994745 | 4.319285007 | -6.66066 | ms | | |
100th percentile latency | index-stats | 18.87320075 | 4.494163208 | -14.37904 | ms | | |
50th percentile service time | index-stats | 1.521472819 | 1.515000127 | -0.00647 | ms | | |
90th percentile service time | index-stats | 1.832026243 | 1.815253124 | -0.01677 | ms | | |
99th percentile service time | index-stats | 2.493222319 | 2.149661649 | -0.34356 | ms | | |
99.9th percentile service time | index-stats | 3.265745808 | 2.558897269 | -0.70685 | ms | | |
100th percentile service time | index-stats | 18.49333011 | 2.714292146 | -15.77904 | ms | | |
error rate | index-stats | 0 | 0 | 0 | % | | |
node 指标统计 | Min Throughput | node-stats | 89.77465925 | 89.9748329 | 0.20017 | ops/s | |
Mean Throughput | node-stats | 89.90322336 | 89.99453842 | 0.09132 | ops/s | | |
Median Throughput | node-stats | 89.92739222 | 89.99716573 | 0.06977 | ops/s | | |
Max Throughput | node-stats | 89.95708367 | 90.01224368 | 0.05516 | ops/s | | |
50th percentile latency | node-stats | 2.864421345 | 2.847921103 | -0.0165 | ms | | |
90th percentile latency | node-stats | 4.03423449 | 4.02287906 | -0.01136 | ms | | |
99th percentile latency | node-stats | 4.780995743 | 4.921176638 | 0.14018 | ms | | |
99.9th percentile latency | node-stats | 11.95199643 | 8.974571229 | -2.97743 | ms | | |
100th percentile latency | node-stats | 19.6932666 | 13.60371523 | -6.08955 | ms | | |
50th percentile service time | node-stats | 2.031991258 | 2.032643184 | 0.00065 | ms | | |
90th percentile service time | node-stats | 2.502718102 | 2.520979941 | 0.01826 | ms | | |
99th percentile service time | node-stats | 3.355697962 | 3.726954078 | 0.37126 | ms | | |
99.9th percentile service time | node-stats | 6.070044631 | 8.64341661 | 2.57337 | ms | | |
100th percentile service time | node-stats | 19.13440693 | 11.63361594 | -7.50079 | ms | | |
error rate | node-stats | 0 | 0 | 0 | % | | |
默认查询，所有文档 score 为1（match_all） | Min Throughput | default | 49.66239789 | 49.88425331 | 0.22186 | ops/s | |
Mean Throughput | default | 49.80337019 | 49.93227136 | 0.1289 | ops/s | | |
Median Throughput | default | 49.8202478 | 49.93857317 | 0.11833 | ops/s | | |
Max Throughput | default | 49.87518669 | 49.9563597 | 0.08117 | ops/s | | |
50th percentile latency | default | 3.394149244 | 3.262675833 | -0.13147 | ms | | |
90th percentile latency | default | 4.578030575 | 4.408122599 | -0.16991 | ms | | |
99th percentile latency | default | 6.253439859 | 4.992298558 | -1.26114 | ms | | |
99.9th percentile latency | default | 19.66198959 | 8.490681676 | -11.17131 | ms | | |
100th percentile latency | default | 20.0197883 | 8.887056261 | -11.13273 | ms | | |
50th percentile service time | default | 2.622524276 | 2.356512472 | -0.26601 | ms | | |
90th percentile service time | default | 3.102052212 | 2.851721831 | -0.25033 | ms | | |
99th percentile service time | default | 4.579989612 | 3.174401047 | -1.40559 | ms | | |
99.9th percentile service time | default | 18.49881567 | 8.051926313 | -10.44689 | ms | | |
100th percentile service time | default | 18.62356346 | 8.21478758 | -10.40878 | ms | | |
error rate | default | 0 | 0 | 0 | % | | |
term 条件查询 | Min Throughput | term | 98.93043451 | 99.71455545 | 0.78412 | ops/s | |
Mean Throughput | term | 99.33413382 | 99.81852777 | 0.48439 | ops/s | | |
Median Throughput | term | 99.38459416 | 99.83007784 | 0.44548 | ops/s | | |
Max Throughput | term | 99.57176235 | 99.88279227 | 0.31103 | ops/s | | |
50th percentile latency | term | 3.250969574 | 3.228969406 | -0.022 | ms | | |
90th percentile latency | term | 3.966032993 | 3.853681684 | -0.11235 | ms | | |
99th percentile latency | term | 10.50691157 | 4.505703636 | -6.00121 | ms | | |
99.9th percentile latency | term | 17.10123536 | 7.033703398 | -10.06753 | ms | | |
100th percentile latency | term | 19.53481138 | 9.737900458 | -9.79691 | ms | | |
50th percentile service time | term | 2.501523588 | 2.488659229 | -0.01286 | ms | | |
90th percentile service time | term | 3.069854062 | 2.982806601 | -0.08705 | ms | | |
99th percentile service time | term | 7.066733902 | 3.509562407 | -3.55717 | ms | | |
99.9th percentile service time | term | 16.17278317 | 6.230151438 | -9.94263 | ms | | |
100th percentile service time | term | 19.29396484 | 8.562799543 | -10.73117 | ms | | |
error rate | term | 0 | 0 | 0 | % | | |
词组查询（query） | Min Throughput | phrase | 109.1666798 | 109.563189 | 0.39651 | ops/s | |
Mean Throughput | phrase | 109.4885892 | 109.726084 | 0.23749 | ops/s | | |
Median Throughput | phrase | 109.531043 | 109.7627623 | 0.23172 | ops/s | | |
Max Throughput | phrase | 109.6776159 | 109.8360981 | 0.15848 | ops/s | | |
50th percentile latency | phrase | 2.736125607 | 2.723197918 | -0.01293 | ms | | |
90th percentile latency | phrase | 3.2537736 | 3.277133778 | 0.02336 | ms | | |
99th percentile latency | phrase | 5.174562978 | 5.252283504 | 0.07772 | ms | | |
99.9th percentile latency | phrase | 17.94986153 | 11.65228278 | -6.29758 | ms | | |
100th percentile latency | phrase | 20.16797382 | 18.0053385 | -2.16264 | ms | | |
50th percentile service time | phrase | 1.983109396 | 1.964103896 | -0.01901 | ms | | |
90th percentile service time | phrase | 2.38582911 | 2.413296979 | 0.02747 | ms | | |
99th percentile service time | phrase | 4.028498856 | 3.426500661 | -0.602 | ms | | |
99.9th percentile service time | phrase | 17.23640091 | 10.3977854 | -6.83862 | ms | | |
100th percentile service time | phrase | 19.54707783 | 17.02223159 | -2.52485 | ms | | |
不带缓存的聚合查询（aggregation） | error rate | phrase | 0 | 0 | 0 | % | |
Min Throughput | country_agg_uncached | 2.996727436 | 2.999315225 | 0.00259 | ops/s | | |
Mean Throughput | country_agg_uncached | 2.997330498 | 2.999446981 | 0.00212 | ops/s | | |
Median Throughput | country_agg_uncached | 2.997367399 | 2.999449396 | 0.00208 | ops/s | | |
Max Throughput | country_agg_uncached | 2.997811835 | 2.999560826 | 0.00175 | ops/s | | |
50th percentile latency | country_agg_uncached | 137.708772 | 138.3623141 | 0.65354 | ms | | |
90th percentile latency | country_agg_uncached | 162.6020876 | 162.0003162 | -0.60177 | ms | | |
99th percentile latency | country_agg_uncached | 196.1384525 | 190.0452044 | -6.09325 | ms | | |
100th percentile latency | country_agg_uncached | 208.7042201 | 205.8009086 | -2.90331 | ms | | |
50th percentile service time | country_agg_uncached | 136.977708 | 137.0970309 | 0.11932 | ms | | |
90th percentile service time | country_agg_uncached | 161.4701347 | 160.9131827 | -0.55695 | ms | | |
99th percentile service time | country_agg_uncached | 195.4892302 | 188.7832217 | -6.70601 | ms | | |
100th percentile service time | country_agg_uncached | 208.3484558 | 204.5730753 | -3.77538 | ms | | |
error rate | country_agg_uncached | 0 | 0 | 0 | % | | |
带缓存的聚合查询（aggregation） | Min Throughput | country_agg_cached | 98.51641526 | 98.62990947 | 0.11349 | ops/s | |
Mean Throughput | country_agg_cached | 98.94635299 | 99.03419609 | 0.08784 | ops/s | | |
Median Throughput | country_agg_cached | 98.99545733 | 99.08216113 | 0.0867 | ops/s | | |
Max Throughput | country_agg_cached | 99.24729184 | 99.31333794 | 0.06605 | ops/s | | |
50th percentile latency | country_agg_cached | 2.211962827 | 2.139798366 | -0.07216 | ms | | |
90th percentile latency | country_agg_cached | 3.517023474 | 3.494661488 | -0.02236 | ms | | |
99th percentile latency | country_agg_cached | 4.158023223 | 4.199306211 | 0.04128 | ms | | |
99.9th percentile latency | country_agg_cached | 9.866942695 | 8.245296748 | -1.62165 | ms | | |
100th percentile latency | country_agg_cached | 18.06280296 | 12.30363548 | -5.75917 | ms | | |
50th percentile service time | country_agg_cached | 1.467651688 | 1.393478829 | -0.07417 | ms | | |
90th percentile service time | country_agg_cached | 1.777389366 | 1.689927001 | -0.08746 | ms | | |
99th percentile service time | country_agg_cached | 2.282693414 | 3.276122157 | 0.99343 | ms | | |
99.9th percentile service time | country_agg_cached | 4.195660669 | 7.769071626 | 3.57341 | ms | | |
100th percentile service time | country_agg_cached | 16.24826528 | 11.39958762 | -4.84868 | ms | | |
error rate | country_agg_cached | 0 | 0 | 0 | % | | |
分页拉取 | Min Throughput | scroll | 20.04421286 | 20.04208025 | -0.00213 | pages/s | |
Mean Throughput | scroll | 20.05368445 | 20.05111381 | -0.00257 | pages/s | | |
Median Throughput | scroll | 20.05292541 | 20.05042813 | -0.0025 | pages/s | | |
Max Throughput | scroll | 20.0660563 | 20.06287951 | -0.00318 | pages/s | | |
50th percentile latency | scroll | 272.1138773 | 259.2352917 | -12.87859 | ms | | |
90th percentile latency | scroll | 290.9470227 | 265.0907522 | -25.85627 | ms | | |
99th percentile latency | scroll | 302.488716 | 284.5098141 | -17.9789 | ms | | |
100th percentile latency | scroll | 303.7193846 | 297.6893578 | -6.03003 | ms | | |
50th percentile service time | scroll | 270.1747189 | 257.1577448 | -13.01697 | ms | | |
90th percentile service time | scroll | 289.0668329 | 263.437889 | -25.62894 | ms | | |
99th percentile service time | scroll | 300.3281443 | 282.0971792 | -18.23097 | ms | | |
100th percentile service time | scroll | 301.2135932 | 296.1045038 | -5.10909 | ms | | |
error rate | scroll | 0 | 0 | 0 | % | | |
脚本查询（使用 expression 脚本） | Min Throughput | expression | 1.500956979 | 1.501441158 | 0.00048 | ops/s | |
Mean Throughput | expression | 1.501160838 | 1.501741788 | 0.00058 | ops/s | | |
Median Throughput | expression | 1.501147998 | 1.501719862 | 0.00057 | ops/s | | |
Max Throughput | expression | 1.501414072 | 1.502131242 | 0.00072 | ops/s | | |
50th percentile latency | expression | 327.3259858 | 295.2159694 | -32.11002 | ms | | |
90th percentile latency | expression | 342.0345129 | 317.3502734 | -24.68424 | ms | | |
99th percentile latency | expression | 372.6446468 | 378.85094 | 6.20629 | ms | | |
100th percentile latency | expression | 396.7165332 | 417.1186928 | 20.40216 | ms | | |
50th percentile service time | expression | 325.855901 | 293.9883978 | -31.8675 | ms | | |
90th percentile service time | expression | 340.6900207 | 316.3654443 | -24.32458 | ms | | |
99th percentile service time | expression | 370.736203 | 377.514769 | 6.77857 | ms | | |
100th percentile service time | expression | 395.4625437 | 415.9661252 | 20.50358 | ms | | |
error rate | expression | 0 | 0 | 0 | % | | |
脚本查询（使用 painless 静态脚本，不动态取字段值） | Min Throughput | painless_static | 1.396916338 | 1.397483837 | 0.00057 | ops/s | |
Mean Throughput | painless_static | 1.397478748 | 1.397943395 | 0.00046 | ops/s | | |
Median Throughput | painless_static | 1.397513941 | 1.397977624 | 0.00046 | ops/s | | |
Max Throughput | painless_static | 1.397920498 | 1.398303982 | 0.00038 | ops/s | | |
50th percentile latency | painless_static | 431.2919118 | 371.348965 | -59.94295 | ms | | |
90th percentile latency | painless_static | 465.1254796 | 391.1945282 | -73.93095 | ms | | |
99th percentile latency | painless_static | 512.2339443 | 437.3164341 | -74.91751 | ms | | |
100th percentile latency | painless_static | 538.9131764 | 465.5702729 | -73.3429 | ms | | |
50th percentile service time | painless_static | 429.9421017 | 369.791389 | -60.15071 | ms | | |
90th percentile service time | painless_static | 463.2926716 | 390.19037 | -73.1023 | ms | | |
99th percentile service time | painless_static | 511.3802825 | 434.9970652 | -76.38322 | ms | | |
100th percentile service time | painless_static | 537.7559569 | 464.3589323 | -73.39702 | ms | | |
error rate | painless_static | 0 | 0 | 0 | % | | |
脚本查询（使用 painless 静态脚本，动态获取字段值） | Min Throughput | painless_dynamic | 1.398724661 | 1.396323104 | -0.0024 | ops/s | |
Mean Throughput | painless_dynamic | 1.398964022 | 1.396996725 | -0.00197 | ops/s | | |
Median Throughput | painless_dynamic | 1.398981831 | 1.397038282 | -0.00194 | ops/s | | |
Max Throughput | painless_dynamic | 1.399149307 | 1.397521303 | -0.00163 | ops/s | | |
50th percentile latency | painless_dynamic | 432.8310895 | 356.7619352 | -76.06915 | ms | | |
90th percentile latency | painless_dynamic | 462.9847418 | 383.0218635 | -79.96288 | ms | | |
99th percentile latency | painless_dynamic | 494.9476089 | 428.2430825 | -66.70453 | ms | | |
100th percentile latency | painless_dynamic | 536.4017347 | 446.9218394 | -89.4799 | ms | | |
50th percentile service time | painless_dynamic | 431.5832192 | 355.6409795 | -75.94224 | ms | | |
90th percentile service time | painless_dynamic | 462.0900041 | 381.7875394 | -80.30246 | ms | | |
99th percentile service time | painless_dynamic | 494.3205597 | 425.9035249 | -68.41703 | ms | | |
100th percentile service time | painless_dynamic | 534.6057713 | 445.0034723 | -89.6023 | ms | | |
error rate | painless_dynamic | 0 | 0 | 0 | % | | |
地理范围查询（基于高斯衰减函数） | Min Throughput | decay_geo_gauss_function_score | 1.001927565 | 1.002114687 | 0.00019 | ops/s | |
Mean Throughput | decay_geo_gauss_function_score | 1.002340802 | 1.002568849 | 0.00023 | ops/s | | |
Median Throughput | decay_geo_gauss_function_score | 1.002308555 | 1.002535216 | 0.00023 | ops/s | | |
Max Throughput | decay_geo_gauss_function_score | 1.002881625 | 1.003158744 | 0.00028 | ops/s | | |
50th percentile latency | decay_geo_gauss_function_score | 387.5082242 | 332.3548282 | -55.1534 | ms | | |
90th percentile latency | decay_geo_gauss_function_score | 397.8741518 | 344.7444949 | -53.12966 | ms | | |
99th percentile latency | decay_geo_gauss_function_score | 407.4444408 | 356.9588375 | -50.4856 | ms | | |
100th percentile latency | decay_geo_gauss_function_score | 409.4531341 | 369.3594299 | -40.0937 | ms | | |
50th percentile service time | decay_geo_gauss_function_score | 386.1244814 | 331.0354254 | -55.08906 | ms | | |
90th percentile service time | decay_geo_gauss_function_score | 396.8515609 | 343.3262392 | -53.52532 | ms | | |
99th percentile service time | decay_geo_gauss_function_score | 406.6675034 | 355.0055938 | -51.66191 | ms | | |
100th percentile service time | decay_geo_gauss_function_score | 407.7369291 | 368.1781357 | -39.55879 | ms | | |
error rate | decay_geo_gauss_function_score | 0 | 0 | 0 | % | | |
地理范围查询（基于高斯衰减函数，且脚本动态获取字段值） | Min Throughput | decay_geo_gauss_script_score | 1.001446744 | 1.001552191 | 0.00011 | ops/s | |
Mean Throughput | decay_geo_gauss_script_score | 1.001755635 | 1.001885174 | 0.00013 | ops/s | | |
Median Throughput | decay_geo_gauss_script_score | 1.001731537 | 1.001860497 | 0.00013 | ops/s | | |
Max Throughput | decay_geo_gauss_script_score | 1.002160032 | 1.002318693 | 0.00016 | ops/s | | |
50th percentile latency | decay_geo_gauss_script_score | 411.4939715 | 334.8041 | -76.68987 | ms | | |
90th percentile latency | decay_geo_gauss_script_score | 429.658707 | 345.120705 | -84.538 | ms | | |
99th percentile latency | decay_geo_gauss_script_score | 453.6645598 | 355.949311 | -97.71525 | ms | | |
100th percentile latency | decay_geo_gauss_script_score | 454.430094 | 358.0469266 | -96.38317 | ms | | |
50th percentile service time | decay_geo_gauss_script_score | 409.7672189 | 333.3149343 | -76.45228 | ms | | |
90th percentile service time | decay_geo_gauss_script_score | 428.3069702 | 343.9684012 | -84.33857 | ms | | |
99th percentile service time | decay_geo_gauss_script_score | 451.8706388 | 354.2061434 | -97.6645 | ms | | |
100th percentile service time | decay_geo_gauss_script_score | 452.8327445 | 356.5891208 | -96.24362 | ms | | |
error rate | decay_geo_gauss_script_score | 0 | 0 | 0 | % | | |
自定义评分函数查询（基于字段值定义函数） | Min Throughput | field_value_function_score | 1.503388048 | 1.503875481 | 0.00049 | ops/s | |
Mean Throughput | field_value_function_score | 1.504118746 | 1.504719285 | 0.0006 | ops/s | | |
Median Throughput | field_value_function_score | 1.504074621 | 1.504659173 | 0.00058 | ops/s | | |
Max Throughput | field_value_function_score | 1.505051463 | 1.505800982 | 0.00075 | ops/s | | |
50th percentile latency | field_value_function_score | 194.2629726 | 134.5724794 | -59.69049 | ms | | |
90th percentile latency | field_value_function_score | 203.7090491 | 150.1895525 | -53.5195 | ms | | |
99th percentile latency | field_value_function_score | 214.6481861 | 166.6002053 | -48.04798 | ms | | |
100th percentile latency | field_value_function_score | 217.926288 | 184.5367327 | -33.38956 | ms | | |
50th percentile service time | field_value_function_score | 192.3880568 | 133.1520383 | -59.23602 | ms | | |
90th percentile service time | field_value_function_score | 202.3297952 | 148.4251454 | -53.90465 | ms | | |
99th percentile service time | field_value_function_score | 213.3810514 | 165.5014484 | -47.8796 | ms | | |
100th percentile service time | field_value_function_score | 215.2935127 | 183.1076834 | -32.18583 | ms | | |
error rate | field_value_function_score | 0 | 0 | 0 | % | | |
自定义评分函数查询（通过脚本动态获取字段值计算评分） | Min Throughput | field_value_script_score | 1.499697369 | 1.500258349 | 0.00056 | ops/s | |
Mean Throughput | field_value_script_score | 1.499757694 | 1.500311799 | 0.00055 | ops/s | | |
Median Throughput | field_value_script_score | 1.499759282 | 1.50030649 | 0.00055 | ops/s | | |
Max Throughput | field_value_script_score | 1.499799232 | 1.500380821 | 0.00058 | ops/s | | |
50th percentile latency | field_value_script_score | 240.0929602 | 174.8193153 | -65.27364 | ms | | |
90th percentile latency | field_value_script_score | 250.0571281 | 188.9238266 | -61.1333 | ms | | |
99th percentile latency | field_value_script_score | 270.1539508 | 215.9618342 | -54.19212 | ms | | |
100th percentile latency | field_value_script_score | 291.1372129 | 229.1083755 | -62.02884 | ms | | |
50th percentile service time | field_value_script_score | 238.8174967 | 173.5835276 | -65.23397 | ms | | |
90th percentile service time | field_value_script_score | 248.7244156 | 187.4786591 | -61.24576 | ms | | |
99th percentile service time | field_value_script_score | 268.9089779 | 214.8508051 | -54.05817 | ms | | |
100th percentile service time | field_value_script_score | 290.2693953 | 228.2811021 | -61.98829 | ms | | |
error rate | field_value_script_score | 0 | 0 | 0 | % | | |
大量 terms 条件查询（query） | Min Throughput | large_terms | 1.101304601 | 1.100700963 | -0.0006 | ops/s | |
Mean Throughput | large_terms | 1.101582867 | 1.100849475 | -0.00073 | ops/s | | |
Median Throughput | large_terms | 1.101561184 | 1.100839723 | -0.00072 | ops/s | | |
Max Throughput | large_terms | 1.101945785 | 1.101043 | -0.0009 | ops/s | | |
50th percentile latency | large_terms | 211.8277326 | 242.00767 | 30.17994 | ms | | |
90th percentile latency | large_terms | 231.8088979 | 252.8580495 | 21.04915 | ms | | |
99th percentile latency | large_terms | 251.2304624 | 265.9718477 | 14.74139 | ms | | |
100th percentile latency | large_terms | 255.3527635 | 269.8639119 | 14.51115 | ms | | |
50th percentile service time | large_terms | 203.8265727 | 233.9178906 | 30.09132 | ms | | |
90th percentile service time | large_terms | 223.8224964 | 245.1530447 | 21.33055 | ms | | |
99th percentile service time | large_terms | 241.849935 | 258.2161737 | 16.36624 | ms | | |
100th percentile service time | large_terms | 246.3486325 | 262.1599194 | 15.81129 | ms | | |
error rate | large_terms | 0 | 0 | 0 | % | | |
大量 terms 条件过滤查询（query、filter） | Min Throughput | large_filtered_terms | 1.102296697 | 1.10245872 | 0.00016 | ops/s | |
Mean Throughput | large_filtered_terms | 1.102784885 | 1.102979701 | 0.00019 | ops/s | | |
Median Throughput | large_filtered_terms | 1.102747397 | 1.102939052 | 0.00019 | ops/s | | |
Max Throughput | large_filtered_terms | 1.103436209 | 1.103668743 | 0.00023 | ops/s | | |
50th percentile latency | large_filtered_terms | 227.9210831 | 243.4361419 | 15.51506 | ms | | |
90th percentile latency | large_filtered_terms | 249.2253724 | 255.9631401 | 6.73777 | ms | | |
99th percentile latency | large_filtered_terms | 263.3142567 | 276.256653 | 12.9424 | ms | | |
100th percentile latency | large_filtered_terms | 265.4732559 | 280.1711811 | 14.69793 | ms | | |
50th percentile service time | large_filtered_terms | 220.1946224 | 235.739741 | 15.54512 | ms | | |
90th percentile service time | large_filtered_terms | 241.1826614 | 248.5632175 | 7.38056 | ms | | |
99th percentile service time | large_filtered_terms | 255.2141531 | 268.2613158 | 13.04716 | ms | | |
100th percentile service time | large_filtered_terms | 256.941474 | 272.5524893 | 15.61102 | ms | | |
error rate | large_filtered_terms | 0 | 0 | 0 | % | | |
大量条件取反查询（query、must not） | Min Throughput | large_prohibited_terms | 1.102827031 | 1.102357904 | -0.00047 | ops/s | |
Mean Throughput | large_prohibited_terms | 1.103422713 | 1.102860804 | -0.00056 | ops/s | | |
Median Throughput | large_prohibited_terms | 1.103376606 | 1.102821884 | -0.00055 | ops/s | | |
Max Throughput | large_prohibited_terms | 1.104211668 | 1.103525116 | -0.00069 | ops/s | | |
50th percentile latency | large_prohibited_terms | 202.5767318 | 232.1078526 | 29.53112 | ms | | |
90th percentile latency | large_prohibited_terms | 220.4174595 | 247.2566163 | 26.83916 | ms | | |
99th percentile latency | large_prohibited_terms | 234.3344817 | 266.7954953 | 32.46101 | ms | | |
100th percentile latency | large_prohibited_terms | 246.6131421 | 268.8084021 | 22.19526 | ms | | |
50th percentile service time | large_prohibited_terms | 193.9010601 | 224.5439067 | 30.64285 | ms | | |
90th percentile service time | large_prohibited_terms | 212.6108234 | 239.8692667 | 27.25844 | ms | | |
99th percentile service time | large_prohibited_terms | 226.4359237 | 258.9916089 | 32.55569 | ms | | |
100th percentile service time | large_prohibited_terms | 238.984541 | 260.8724702 | 21.88793 | ms | | |
error rate | large_prohibited_terms | 0 | 0 | 0 | % | | |
降序排序查询 | Min Throughput | desc_sort_population | 1.504037884 | 1.504180086 | 0.00014 | ops/s | |
Mean Throughput | desc_sort_population | 1.504907329 | 1.505087394 | 0.00018 | ops/s | | |
Median Throughput | desc_sort_population | 1.504841628 | 1.505025118 | 0.00018 | ops/s | | |
Max Throughput | desc_sort_population | 1.506034196 | 1.506249517 | 0.00022 | ops/s | | |
50th percentile latency | desc_sort_population | 61.13778474 | 54.50106226 | -6.63672 | ms | | |
90th percentile latency | desc_sort_population | 73.92849587 | 69.35394919 | -4.57455 | ms | | |
99th percentile latency | desc_sort_population | 85.77715084 | 86.2006122 | 0.42346 | ms | | |
100th percentile latency | desc_sort_population | 85.84200498 | 87.74290979 | 1.9009 | ms | | |
50th percentile service time | desc_sort_population | 59.92501229 | 53.58439684 | -6.34062 | ms | | |
90th percentile service time | desc_sort_population | 72.30911367 | 68.09855495 | -4.21056 | ms | | |
99th percentile service time | desc_sort_population | 84.09957183 | 84.57749833 | 0.47793 | ms | | |
100th percentile service time | desc_sort_population | 84.19063408 | 85.95814556 | 1.76751 | ms | | |
error rate | desc_sort_population | 0 | 0 | 0 | % | | |
升序排序查询 | Min Throughput | asc_sort_population | 1.504247142 | 1.504528234 | 0.00028 | ops/s | |
Mean Throughput | asc_sort_population | 1.505166062 | 1.505508302 | 0.00034 | ops/s | | |
Median Throughput | asc_sort_population | 1.505099341 | 1.505440428 | 0.00034 | ops/s | | |
Max Throughput | asc_sort_population | 1.506349989 | 1.506767598 | 0.00042 | ops/s | | |
50th percentile latency | asc_sort_population | 63.16776341 | 62.82690261 | -0.34086 | ms | | |
90th percentile latency | asc_sort_population | 78.09764324 | 75.61749704 | -2.48015 | ms | | |
99th percentile latency | asc_sort_population | 87.33172638 | 84.58683862 | -2.74489 | ms | | |
100th percentile latency | asc_sort_population | 87.89979294 | 85.19899659 | -2.7008 | ms | | |
50th percentile service time | asc_sort_population | 61.90986466 | 61.71039958 | -0.19947 | ms | | |
90th percentile service time | asc_sort_population | 76.55056985 | 74.45018981 | -2.10038 | ms | | |
99th percentile service time | asc_sort_population | 85.81453795 | 83.30245529 | -2.51208 | ms | | |
100th percentile service time | asc_sort_population | 86.60695888 | 84.05557927 | -2.55138 | ms | | |
error rate | asc_sort_population | 0 | 0 | 0 | % | | |
升序排序后 after 跳转查询 | Min Throughput | asc_sort_with_after_population | 1.503017792 | 1.503506494 | 0.00049 | ops/s | |
Mean Throughput | asc_sort_with_after_population | 1.503667166 | 1.504271472 | 0.0006 | ops/s | | |
Median Throughput | asc_sort_with_after_population | 1.503620246 | 1.504214876 | 0.00059 | ops/s | | |
Max Throughput | asc_sort_with_after_population | 1.504506304 | 1.505248472 | 0.00074 | ops/s | | |
50th percentile latency | asc_sort_with_after_population | 79.49512405 | 81.97531058 | 2.48019 | ms | | |
90th percentile latency | asc_sort_with_after_population | 94.07415418 | 97.05960844 | 2.98545 | ms | | |
99th percentile latency | asc_sort_with_after_population | 115.1407234 | 110.1778366 | -4.96289 | ms | | |
100th percentile latency | asc_sort_with_after_population | 117.4867153 | 115.6357806 | -1.85093 | ms | | |
50th percentile service time | asc_sort_with_after_population | 78.12653808 | 80.56232799 | 2.43579 | ms | | |
90th percentile service time | asc_sort_with_after_population | 92.57791536 | 96.00112103 | 3.42321 | ms | | |
99th percentile service time | asc_sort_with_after_population | 113.538067 | 108.2517642 | -5.2863 | ms | | |
100th percentile service time | asc_sort_with_after_population | 116.0558164 | 114.5531256 | -1.50269 | ms | | |
error rate | asc_sort_with_after_population | 0 | 0 | 0 | % | | |
高基字段降序排序查询（基于 DistanceFeatureQuery 快速取 topK） | Min Throughput | desc_sort_geonameid | 6.011806976 | 6.013534032 | 0.00173 | ops/s | |
Mean Throughput | desc_sort_geonameid | 6.014040004 | 6.016121536 | 0.00208 | ops/s | | |
Median Throughput | desc_sort_geonameid | 6.013860893 | 6.015844404 | 0.00198 | ops/s | | |
Max Throughput | desc_sort_geonameid | 6.016975785 | 6.019491653 | 0.00252 | ops/s | | |
50th percentile latency | desc_sort_geonameid | 8.037513588 | 6.896098144 | -1.14142 | ms | | |
90th percentile latency | desc_sort_geonameid | 8.790209144 | 7.481213845 | -1.309 | ms | | |
99th percentile latency | desc_sort_geonameid | 20.16597 | 7.890859395 | -12.27511 | ms | | |
100th percentile latency | desc_sort_geonameid | 22.69194461 | 8.130467497 | -14.56148 | ms | | |
50th percentile service time | desc_sort_geonameid | 7.199986372 | 6.043605972 | -1.15638 | ms | | |
90th percentile service time | desc_sort_geonameid | 7.634483464 | 6.330675445 | -1.30381 | ms | | |
99th percentile service time | desc_sort_geonameid | 18.95111335 | 6.674837489 | -12.27628 | ms | | |
100th percentile service time | desc_sort_geonameid | 22.06934988 | 6.795545109 | -15.2738 | ms | | |
error rate | desc_sort_geonameid | 0 | 0 | 0 | % | | |
高基字段降序排序 after 跳转查询 | Min Throughput | desc_sort_with_after_geonameid | 6.003999251 | 6.002224615 | -0.00177 | ops/s | |
Mean Throughput | desc_sort_with_after_geonameid | 6.00483332 | 6.002715504 | -0.00212 | ops/s | | |
Median Throughput | desc_sort_with_after_geonameid | 6.004831591 | 6.002684836 | -0.00215 | ops/s | | |
Max Throughput | desc_sort_with_after_geonameid | 6.005864935 | 6.003257919 | -0.00261 | ops/s | | |
50th percentile latency | desc_sort_with_after_geonameid | 64.12782287 | 69.3480419 | 5.22022 | ms | | |
90th percentile latency | desc_sort_with_after_geonameid | 79.63361973 | 85.98741582 | 6.3538 | ms | | |
99th percentile latency | desc_sort_with_after_geonameid | 87.09606319 | 91.30932659 | 4.21326 | ms | | |
100th percentile latency | desc_sort_with_after_geonameid | 88.47462852 | 91.78488795 | 3.31026 | ms | | |
50th percentile service time | desc_sort_with_after_geonameid | 63.23770666 | 68.51645093 | 5.27874 | ms | | |
90th percentile service time | desc_sort_with_after_geonameid | 78.83979175 | 85.22403594 | 6.38424 | ms | | |
99th percentile service time | desc_sort_with_after_geonameid | 86.525729 | 90.76162191 | 4.23589 | ms | | |
100th percentile service time | desc_sort_with_after_geonameid | 87.29847241 | 91.3709281 | 4.07246 | ms | | |
error rate | desc_sort_with_after_geonameid | 0 | 0 | 0 | % | | |
高基字段升序排序查询（基于 DistanceFeatureQuery 快速取 topK） | Min Throughput | asc_sort_geonameid | 6.018840993 | 6.018470998 | -0.00037 | ops/s | |
Mean Throughput | asc_sort_geonameid | 6.022518134 | 6.022078364 | -0.00044 | ops/s | | |
Median Throughput | asc_sort_geonameid | 6.022245684 | 6.021816641 | -0.00043 | ops/s | | |
Max Throughput | asc_sort_geonameid | 6.027178807 | 6.026594943 | -0.00058 | ops/s | | |
50th percentile latency | asc_sort_geonameid | 7.024060003 | 9.012220893 | 1.98816 | ms | | |
90th percentile latency | asc_sort_geonameid | 7.69297732 | 9.680523816 | 1.98755 | ms | | |
99th percentile latency | asc_sort_geonameid | 20.44826921 | 11.18117133 | -9.2671 | ms | | |
100th percentile latency | asc_sort_geonameid | 21.87036537 | 11.28741447 | -10.58295 | ms | | |
50th percentile service time | asc_sort_geonameid | 6.123304367 | 8.064015303 | 1.94071 | ms | | |
90th percentile service time | asc_sort_geonameid | 6.746383384 | 8.737695683 | 1.99131 | ms | | |
99th percentile service time | asc_sort_geonameid | 19.78318544 | 10.16213525 | -9.62105 | ms | | |
100th percentile service time | asc_sort_geonameid | 21.25467733 | 10.39039157 | -10.86429 | ms | | |
error rate | asc_sort_geonameid | 0 | 0 | 0 | % | | |
高基字段升序排序 after 跳转查询 | Min Throughput | asc_sort_with_after_geonameid | 6.013236842 | 6.013266563 | 0.00003 | ops/s | |
Mean Throughput | asc_sort_with_after_geonameid | 6.015849171 | 6.015858176 | 0.00001 | ops/s | | |
Median Throughput | asc_sort_with_after_geonameid | 6.015618744 | 6.015641333 | 0.00002 | ops/s | | |
Max Throughput | asc_sort_with_after_geonameid | 6.019167352 | 6.01911523 | -0.00005 | ops/s | | |
50th percentile latency | asc_sort_with_after_geonameid | 60.27546292 | 64.3463349 | 4.07087 | ms | | |
90th percentile latency | asc_sort_with_after_geonameid | 78.63363056 | 85.38805693 | 6.75443 | ms | | |
99th percentile latency | asc_sort_with_after_geonameid | 89.31191583 | 91.7664034 | 2.45449 | ms | | |
100th percentile latency | asc_sort_with_after_geonameid | 90.85853212 | 91.9917766 | 1.13324 | ms | | |
50th percentile service time | asc_sort_with_after_geonameid | 59.692265 | 63.68059153 | 3.98833 | ms | | |
90th percentile service time | asc_sort_with_after_geonameid | 78.16235274 | 84.53184282 | 6.36949 | ms | | |
99th percentile service time | asc_sort_with_after_geonameid | 88.15484255 | 91.29356634 | 3.13872 | ms | | |
100th percentile service time | asc_sort_with_after_geonameid | 89.73695803 | 91.64701309 | 1.91006 | ms | | |
error rate | asc_sort_with_after_geonameid | 0 | 0 | 0 | % | | |
文档反馈