本文提供8核32G SSD 200G 3个节点的腾讯云 Elasticsearch Service 集群的性能指标。
说明：
对比8核32GB SSD 200G 3节点的腾讯云 ES 和社区版本的集群性能指标，腾讯云 ES 各方面性能均有一定优势。主要得益于腾讯云 ES 在底层存储模型优化（时序合并、冷分片持续合并）、查询执行计划优化（高效剪枝、缓存优化）、内置场景模板、自研 JDK、GC 参数调优等方面的优化，详见 内核增强。
geonames/7.10.1/8核32GB
说明 | Metric | Unit | Task | 腾讯云 ES | 社区版本 | 差异值(社区版-ES) | 差异百分比(差异值/社区版) | 效果比较 |
写入总耗时 | Cumulative indexing time of primary shards | min | - | 14.2567 | 15.4834 | 1.2267 | 7.92% | 优 |
GC 总次数、耗时统计 | Total Young Gen GC time | s | - | 3.544 | 17.73 | 14.186 | 80.01% | 优 |
| Total Young Gen GC count | 次 | - | 447 | 1084 | 637 | 58.76% | 优 |
| Total Old Gen GC time | s | - | 0 | 0 | 0 | - | - |
| Total Old Gen GC count | 次 | - | 0 | 0 | 0 | - | - |
存储大小 | Store size | GB | - | 2.59725 | 3.07604 | 0.47879 | 15.57% | 优 |
堆内存使用量 | Heap used for segments | MB | - | 0.534325 | 0.766575 | 0.23225 | 30.30% | 优 |
| Heap used for doc values | MB | - | 0.0507355 | 0.0315514 | -0.0191841 | -60.80% | 劣 |
| Heap used for terms | MB | - | 0.370026 | 0.601349 | 0.231323 | 38.47% | 优 |
| Heap used for norms | MB | - | 0.0396729 | 0.0817871 | 0.0421142 | 51.49% | 优 |
| Heap used for points | MB | - | 0 | 0 | 0 | - | - |
| Heap used for stored fields | MB | - | 0.0119553 | 0.0518875 | 0.0399322 | 76.96% | 优 |
segment 总数量 | Segment count | 次 | - | 7 | 105 | 98 | 93.33% | 优 |
写入吞吐、耗时统计 | Min Throughput | docs/s | index-append | 153730 | 132776 | -20954 | -15.78% | 优 |
| Median Throughput | docs/s | index-append | 159765 | 142851 | -16914 | -11.84% | 优 |
| Max Throughput | docs/s | index-append | 162791 | 148669 | -14122 | -9.50% | 优 |
| 50th percentile latency | ms | index-append | 130.877 | 151.656 | 20.779 | 13.70% | 优 |
| 90th percentile latency | ms | index-append | 162.969 | 187.779 | 24.81 | 13.21% | 优 |
| 99th percentile latency | ms | index-append | 181.428 | 240.11 | 58.682 | 24.44% | 优 |
| 100th percentile latency | ms | index-append | 225.98 | 285.137 | 59.157 | 20.75% | 优 |
| 50th percentile service time | ms | index-append | 130.877 | 151.656 | 20.779 | 13.70% | 优 |
| 90th percentile service time | ms | index-append | 162.969 | 187.779 | 24.81 | 13.21% | 优 |
| 99th percentile service time | ms | index-append | 181.428 | 240.11 | 58.682 | 24.44% | 优 |
| 100th percentile service time | ms | index-append | 225.98 | 285.137 | 59.157 | 20.75% | 优 |
| error rate | % | index-append | 0 | 0 | 0 | - | - |
index 指标统计 | Min Throughput | ops/s | index-stats | 90.05 | 90.05 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | index-stats | 90.06 | 90.07 | 0.01 | 0.01% | 优 |
| Max Throughput | ops/s | index-stats | 90.12 | 90.13 | 0.01 | 0.01% | 优 |
| 50th percentile latency | ms | index-stats | 2.76736 | 2.65897 | -0.10839 | -4.08% | 劣 |
| 90th percentile latency | ms | index-stats | 3.58235 | 3.48937 | -0.09298 | -2.66% | 劣 |
| 99th percentile latency | ms | index-stats | 3.95798 | 3.89304 | -0.06494 | -1.67% | 劣 |
| 99.9th percentile latency | ms | index-stats | 4.39377 | 9.07619 | 4.68242 | 51.59% | 优 |
| 100th percentile latency | ms | index-stats | 9.00375 | 18.123 | 9.11925 | 50.32% | 优 |
| 50th percentile service time | ms | index-stats | 1.57744 | 1.45244 | -0.125 | -8.61% | 劣 |
| 90th percentile service time | ms | index-stats | 1.8317 | 1.66173 | -0.16997 | -10.23% | 劣 |
| 99th percentile service time | ms | index-stats | 2.0752 | 1.94615 | -0.12905 | -6.63% | 劣 |
| 99.9th percentile service time | ms | index-stats | 2.24891 | 2.64752 | 0.39861 | 15.06% | 优 |
| 100th percentile service time | ms | index-stats | 2.31078 | 16.5015 | 14.19072 | 86.00% | 优 |
| error rate | % | index-stats | 0 | 0 | 0 | - | - |
node 指标统计 | Min Throughput | ops/s | node-stats | 90.06 | 90.06 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | node-stats | 90.12 | 90.12 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | node-stats | 90.36 | 90.39 | 0.03 | 0.03% | 优 |
| 50th percentile latency | ms | node-stats | 2.9754 | 2.93862 | -0.03678 | -1.25% | 劣 |
| 90th percentile latency | ms | node-stats | 4.07929 | 4.0283 | -0.05099 | -1.27% | 劣 |
| 99th percentile latency | ms | node-stats | 5.0754 | 4.78946 | -0.28594 | -5.97% | 劣 |
| 99.9th percentile latency | ms | node-stats | 6.53613 | 17.0702 | 10.53407 | 61.71% | 优 |
| 100th percentile latency | ms | node-stats | 6.93454 | 24.7832 | 17.84866 | 72.02% | 优 |
| 50th percentile service time | ms | node-stats | 2.23841 | 2.17713 | -0.06128 | -2.81% | 劣 |
| 90th percentile service time | ms | node-stats | 2.65367 | 2.62466 | -0.02901 | -1.11% | 劣 |
| 99th percentile service time | ms | node-stats | 3.92073 | 3.93043 | 0.0097 | 0.25% | 优 |
| 99.9th percentile service time | ms | node-stats | 4.92842 | 4.48172 | -0.4467 | -9.97% | 劣 |
| 100th percentile service time | ms | node-stats | 5.92757 | 23.7859 | 17.85833 | 75.08% | 优 |
| error rate | % | node-stats | 0 | 0 | 0 | - | - |
默认查询，所有文档 score 为1（match_all） | Min Throughput | ops/s | default | 50.03 | 50.03 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | default | 50.04 | 50.04 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | default | 50.08 | 50.08 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | default | 3.53894 | 3.40291 | -0.13603 | -4.00% | 劣 |
| 90th percentile latency | ms | default | 4.11403 | 4.57279 | 0.45876 | 10.03% | 优 |
| 99th percentile latency | ms | default | 4.92737 | 5.37426 | 0.44689 | 8.32% | 优 |
| 99.9th percentile latency | ms | default | 5.74037 | 25.897 | 20.15663 | 77.83% | 优 |
| 100th percentile latency | ms | default | 7.32557 | 27.2085 | 19.88293 | 73.08% | 优 |
| 50th percentile service time | ms | default | 2.7831 | 2.67459 | -0.10851 | -4.06% | 劣 |
| 90th percentile service time | ms | default | 3.17322 | 2.97996 | -0.19326 | -6.49% | 劣 |
| 99th percentile service time | ms | default | 3.77477 | 3.38794 | -0.38683 | -11.42% | 劣 |
| 99.9th percentile service time | ms | default | 4.19186 | 25.1729 | 20.98104 | 83.35% | 优 |
| 100th percentile service time | ms | default | 6.58243 | 26.1698 | 19.58737 | 74.85% | 优 |
| error rate | % | default | 0 | 0 | 0 | - | - |
term 条件查询 | Min Throughput | ops/s | term | 99.66 | 100.05 | 0.39 | 0.39% | 优 |
| Median Throughput | ops/s | term | 100.07 | 100.08 | 0.01 | 0.01% | 优 |
| Max Throughput | ops/s | term | 100.11 | 100.15 | 0.04 | 0.04% | 优 |
| 50th percentile latency | ms | term | 2.83987 | 2.72933 | -0.11054 | -4.05% | 劣 |
| 90th percentile latency | ms | term | 3.32569 | 3.16805 | -0.15764 | -4.98% | 劣 |
| 99th percentile latency | ms | term | 3.96055 | 3.66753 | -0.29302 | -7.99% | 劣 |
| 99.9th percentile latency | ms | term | 4.33961 | 9.62272 | 5.28311 | 54.90% | 优 |
| 100th percentile latency | ms | term | 5.70421 | 16.4559 | 10.75169 | 65.34% | 优 |
| 50th percentile service time | ms | term | 2.08935 | 2.00258 | -0.08677 | -4.33% | 劣 |
| 90th percentile service time | ms | term | 2.53284 | 2.24384 | -0.289 | -12.88% | 劣 |
| 99th percentile service time | ms | term | 2.99484 | 2.64135 | -0.35349 | -13.38% | 劣 |
| 99.9th percentile service time | ms | term | 3.37709 | 6.10442 | 2.72733 | 44.68% | 优 |
| 100th percentile service time | ms | term | 5.24029 | 15.8044 | 10.56411 | 66.84% | 优 |
| error rate | % | term | 0 | 0 | 0 | - | - |
词组查询（query） | Min Throughput | ops/s | phrase | 110.04 | 110.06 | 0.02 | 0.02% | 优 |
| Median Throughput | ops/s | phrase | 110.08 | 110.07 | -0.01 | -0.01% | 优 |
| Max Throughput | ops/s | phrase | 110.11 | 110.11 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | phrase | 2.74088 | 2.89197 | 0.15109 | 5.22% | 优 |
| 90th percentile latency | ms | phrase | 3.30207 | 3.39345 | 0.09138 | 2.69% | 优 |
| 99th percentile latency | ms | phrase | 4.8127 | 8.22389 | 3.41119 | 41.48% | 优 |
| 99.9th percentile latency | ms | phrase | 5.57204 | 25.9978 | 20.42576 | 78.57% | 优 |
| 100th percentile latency | ms | phrase | 6.54587 | 27.9116 | 21.36573 | 76.55% | 优 |
| 50th percentile service time | ms | phrase | 1.98839 | 2.18937 | 0.20098 | 9.18% | 优 |
| 90th percentile service time | ms | phrase | 2.41365 | 2.51158 | 0.09793 | 3.90% | 优 |
| 99th percentile service time | ms | phrase | 4.00121 | 3.24254 | -0.75867 | -23.40% | 劣 |
| 99.9th percentile service time | ms | phrase | 4.70793 | 25.7224 | 21.01447 | 81.70% | 优 |
| 100th percentile service time | ms | phrase | 5.67829 | 27.4831 | 21.80481 | 79.34% | 优 |
| error rate | % | phrase | 0 | 0 | 0 | - | - |
不带缓存的聚合查询（aggregation） | Min Throughput | ops/s | country_agg_uncached | 3.6 | 3.6 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | country_agg_uncached | 3.61 | 3.61 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | country_agg_uncached | 3.61 | 3.61 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | country_agg_uncached | 130.314 | 162.61 | 32.296 | 19.86% | 优 |
| 90th percentile latency | ms | country_agg_uncached | 147.567 | 176.4 | 28.833 | 16.35% | 优 |
| 99th percentile latency | ms | country_agg_uncached | 165.174 | 184.368 | 19.194 | 10.41% | 优 |
| 100th percentile latency | ms | country_agg_uncached | 174.015 | 269.972 | 95.957 | 35.54% | 优 |
| 50th percentile service time | ms | country_agg_uncached | 129.186 | 161.723 | 32.537 | 20.12% | 优 |
| 90th percentile service time | ms | country_agg_uncached | 146.921 | 175.819 | 28.898 | 16.44% | 优 |
| 99th percentile service time | ms | country_agg_uncached | 164.579 | 183.7 | 19.121 | 10.41% | 优 |
| 100th percentile service time | ms | country_agg_uncached | 172.827 | 269.339 | 96.512 | 35.83% | 优 |
| error rate | % | country_agg_uncached | 0 | 0 | 0 | - | - |
带缓存的聚合查询（aggregation） | Min Throughput | ops/s | country_agg_cached | 100.04 | 100.03 | -0.01 | -0.01% | 优 |
| Median Throughput | ops/s | country_agg_cached | 100.05 | 100.06 | 0.01 | 0.01% | 优 |
| Max Throughput | ops/s | country_agg_cached | 100.08 | 100.08 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | country_agg_cached | 2.29531 | 2.10251 | -0.1928 | -9.17% | 劣 |
| 90th percentile latency | ms | country_agg_cached | 3.57418 | 3.49217 | -0.08201 | -2.35% | 劣 |
| 99th percentile latency | ms | country_agg_cached | 3.91685 | 3.74707 | -0.16978 | -4.53% | 劣 |
| 99.9th percentile latency | ms | country_agg_cached | 4.19749 | 4.05757 | -0.13992 | -3.45% | 劣 |
| 100th percentile latency | ms | country_agg_cached | 4.51842 | 4.63039 | 0.11197 | 2.42% | 优 |
| 50th percentile service time | ms | country_agg_cached | 1.57861 | 1.40612 | -0.17249 | -12.27% | 劣 |
| 90th percentile service time | ms | country_agg_cached | 1.89111 | 1.58866 | -0.30245 | -19.04% | 劣 |
| 99th percentile service time | ms | country_agg_cached | 2.19488 | 1.74583 | -0.44905 | -25.72% | 劣 |
| 99.9th percentile service time | ms | country_agg_cached | 3.42563 | 3.46564 | 0.04001 | 1.15% | 优 |
| 100th percentile service time | ms | country_agg_cached | 4.28971 | 3.97654 | -0.31317 | -7.88% | 优 |
| error rate | % | country_agg_cached | 0 | 0 | 0 | - | - |
分页拉取 | Min Throughput | pages/s | scroll | 20.04 | 20.04 | 0 | 0.00% | 优 |
| Median Throughput | pages/s | scroll | 20.05 | 20.04 | -0.01 | -0.05% | 优 |
| Max Throughput | pages/s | scroll | 20.06 | 20.06 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | scroll | 538.421 | 556.145 | 17.724 | 3.19% | 优 |
| 90th percentile latency | ms | scroll | 543.566 | 573.903 | 30.337 | 5.29% | 优 |
| 99th percentile latency | ms | scroll | 582.263 | 585.602 | 3.339 | 0.57% | 优 |
| 100th percentile latency | ms | scroll | 584.75 | 587.091 | 2.341 | 0.40% | 优 |
| 50th percentile service time | ms | scroll | 537.068 | 554.325 | 17.257 | 3.11% | 优 |
| 90th percentile service time | ms | scroll | 542.428 | 572.308 | 29.88 | 5.22% | 优 |
| 99th percentile service time | ms | scroll | 580.372 | 583.701 | 3.329 | 0.57% | 优 |
| 100th percentile service time | ms | scroll | 583.612 | 584.764 | 1.152 | 0.20% | 优 |
| error rate | % | scroll | 0 | 0 | 0 | - | - |
脚本查询（使用 expression 脚本） | Min Throughput | ops/s | expression | 2 | 2 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | expression | 2 | 2 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | expression | 2 | 2 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | expression | 265.631 | 277.403 | 11.772 | 4.24% | 优 |
| 90th percentile latency | ms | expression | 287.121 | 299.5 | 12.379 | 4.13% | 优 |
| 99th percentile latency | ms | expression | 311.788 | 434.462 | 122.674 | 28.24% | 优 |
| 100th percentile latency | ms | expression | 391.745 | 446.275 | 54.53 | 12.22% | 优 |
| 50th percentile service time | ms | expression | 264.462 | 276.536 | 12.074 | 4.37% | 优 |
| 90th percentile service time | ms | expression | 285.113 | 298.434 | 13.321 | 4.46% | 优 |
| 99th percentile service time | ms | expression | 310.991 | 434.014 | 123.023 | 28.35% | 优 |
| 100th percentile service time | ms | expression | 390.33 | 445.661 | 55.331 | 12.42% | 优 |
| 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 | ops/s | painless_static | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | painless_static | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | painless_static | 1.5 | 1.5 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | painless_static | 337.96 | 364.636 | 26.676 | 7.32% | 优 |
| 90th percentile latency | ms | painless_static | 358.738 | 383.504 | 24.766 | 6.46% | 优 |
| 99th percentile latency | ms | painless_static | 375.017 | 459.529 | 84.512 | 18.39% | 优 |
| 100th percentile latency | ms | painless_static | 395.417 | 557.708 | 162.291 | 29.10% | 优 |
| 50th percentile service time | ms | painless_static | 337.111 | 363.478 | 26.367 | 7.25% | 优 |
| 90th percentile service time | ms | painless_static | 357.771 | 382.45 | 24.679 | 6.45% | 优 |
| 99th percentile service time | ms | painless_static | 374.121 | 458.343 | 84.222 | 18.38% | 优 |
| 100th percentile service time | ms | painless_static | 394.632 | 556.401 | 161.769 | 29.07% | 优 |
| 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 | ops/s | painless_dynamic | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | painless_dynamic | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | painless_dynamic | 1.5 | 1.5 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | painless_dynamic | 334.684 | 365.354 | 30.67 | 8.39% | 优 |
| 90th percentile latency | ms | painless_dynamic | 354.406 | 390.406 | 36 | 9.22% | 优 |
| 99th percentile latency | ms | painless_dynamic | 377.214 | 552.417 | 175.203 | 31.72% | 优 |
| 100th percentile latency | ms | painless_dynamic | 381.276 | 554.12 | 172.844 | 31.19% | 优 |
| 50th percentile service time | ms | painless_dynamic | 333.654 | 364.353 | 30.699 | 8.43% | 优 |
| 90th percentile service time | ms | painless_dynamic | 353.246 | 389.419 | 36.173 | 9.29% | 优 |
| 99th percentile service time | ms | painless_dynamic | 375.956 | 551.619 | 175.663 | 31.84% | 优 |
| 100th percentile service time | ms | painless_dynamic | 379.75 | 553.208 | 173.458 | 31.35% | 优 |
| error rate | % | painless_dynamic | 0 | 0 | 0 | - | - |
地理范围查询（基于高斯衰减函数） | Min Throughput | ops/s | decay_geo_gauss_function_score | 1 | 1 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | decay_geo_gauss_function_score | 1 | 1 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | decay_geo_gauss_function_score | 1 | 1 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | decay_geo_gauss_function_score | 327.972 | 327.664 | -0.308 | -0.09% | 劣 |
| 90th percentile latency | ms | decay_geo_gauss_function_score | 336.979 | 344.367 | 7.388 | 2.15% | 优 |
| 99th percentile latency | ms | decay_geo_gauss_function_score | 343.562 | 417.386 | 73.824 | 17.69% | 优 |
| 100th percentile latency | ms | decay_geo_gauss_function_score | 344.135 | 420.996 | 76.861 | 18.26% | 优 |
| 50th percentile service time | ms | decay_geo_gauss_function_score | 326.554 | 326.503 | -0.051 | -0.02% | 劣 |
| 90th percentile service time | ms | decay_geo_gauss_function_score | 336.053 | 343.32 | 7.267 | 2.12% | 优 |
| 99th percentile service time | ms | decay_geo_gauss_function_score | 342.151 | 416.72 | 74.569 | 17.89% | 优 |
| 100th percentile service time | ms | decay_geo_gauss_function_score | 342.843 | 419.488 | 76.645 | 18.27% | 优 |
| 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 | ops/s | decay_geo_gauss_script_score | 1 | 1 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | decay_geo_gauss_script_score | 1 | 1 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | decay_geo_gauss_script_score | 1 | 1 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | decay_geo_gauss_script_score | 341.152 | 344.421 | 3.269 | 0.95% | 优 |
| 90th percentile latency | ms | decay_geo_gauss_script_score | 349.94 | 371.575 | 21.635 | 5.82% | 优 |
| 99th percentile latency | ms | decay_geo_gauss_script_score | 354.76 | 420.658 | 65.898 | 15.67% | 优 |
| 100th percentile latency | ms | decay_geo_gauss_script_score | 364.169 | 438.541 | 74.372 | 16.96% | 优 |
| 50th percentile service time | ms | decay_geo_gauss_script_score | 339.967 | 342.961 | 2.994 | 0.87% | 优 |
| 90th percentile service time | ms | decay_geo_gauss_script_score | 348.493 | 370.146 | 21.653 | 5.85% | 优 |
| 99th percentile service time | ms | decay_geo_gauss_script_score | 353.559 | 418.767 | 65.208 | 15.57% | 优 |
| 100th percentile service time | ms | decay_geo_gauss_script_score | 362.748 | 437.195 | 74.447 | 17.03% | 优 |
| 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 | ops/s | field_value_function_score | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | field_value_function_score | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | field_value_function_score | 1.51 | 1.51 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | field_value_function_score | 120.538 | 129.323 | 8.785 | 6.79% | 优 |
| 90th percentile latency | ms | field_value_function_score | 137.702 | 152.985 | 15.283 | 9.99% | 优 |
| 99th percentile latency | ms | field_value_function_score | 147.851 | 185.772 | 37.921 | 20.41% | 优 |
| 100th percentile latency | ms | field_value_function_score | 169.961 | 186.405 | 16.444 | 8.82% | 优 |
| 50th percentile service time | ms | field_value_function_score | 119.159 | 128.146 | 8.987 | 7.01% | 优 |
| 90th percentile service time | ms | field_value_function_score | 136.338 | 151.555 | 15.217 | 10.04% | 优 |
| 99th percentile service time | ms | field_value_function_score | 146.981 | 184.934 | 37.953 | 20.52% | 优 |
| 100th percentile service time | ms | field_value_function_score | 168.964 | 185.269 | 16.305 | 8.80% | 优 |
| 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 | ops/s | field_value_script_score | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | field_value_script_score | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | field_value_script_score | 1.51 | 1.51 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | field_value_script_score | 168.069 | 171.907 | 3.838 | 2.23% | 优 |
| 90th percentile latency | ms | field_value_script_score | 178.933 | 184.467 | 5.534 | 3.00% | 优 |
| 99th percentile latency | ms | field_value_script_score | 196.982 | 200.137 | 3.155 | 1.58% | 优 |
| 100th percentile latency | ms | field_value_script_score | 198.722 | 206.461 | 7.739 | 3.75% | 优 |
| 50th percentile service time | ms | field_value_script_score | 166.827 | 170.653 | 3.826 | 2.24% | 优 |
| 90th percentile service time | ms | field_value_script_score | 177.869 | 183.324 | 5.455 | 2.98% | 优 |
| 99th percentile service time | ms | field_value_script_score | 195.586 | 199.107 | 3.521 | 1.77% | 优 |
| 100th percentile service time | ms | field_value_script_score | 197.054 | 205.323 | 8.269 | 4.03% | 优 |
| error rate | % | field_value_script_score | 0 | 0 | 0 | - | - |
大量 terms 条件查询（query） | Min Throughput | ops/s | large_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | large_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | large_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | large_terms | 241.322 | 597.464 | 356.142 | 59.61% | 优 |
| 90th percentile latency | ms | large_terms | 252.637 | 600.941 | 348.304 | 57.96% | 优 |
| 99th percentile latency | ms | large_terms | 265.807 | 749.895 | 484.088 | 64.55% | 优 |
| 100th percentile latency | ms | large_terms | 272.611 | 751.444 | 478.833 | 63.72% | 优 |
| 50th percentile service time | ms | large_terms | 233.129 | 589.829 | 356.7 | 60.48% | 优 |
| 90th percentile service time | ms | large_terms | 244.494 | 593.562 | 349.068 | 58.81% | 优 |
| 99th percentile service time | ms | large_terms | 258.894 | 742.68 | 483.786 | 65.14% | 优 |
| 100th percentile service time | ms | large_terms | 264.352 | 743.9 | 479.548 | 64.46% | 优 |
| error rate | % | large_terms | 0 | 0 | 0 | - | - |
大量 terms 条件过滤查询（query、filter） | Min Throughput | ops/s | large_filtered_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | large_filtered_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | large_filtered_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | large_filtered_terms | 233.192 | 596.04 | 362.848 | 60.88% | 优 |
| 90th percentile latency | ms | large_filtered_terms | 241.102 | 604.321 | 363.219 | 60.10% | 优 |
| 99th percentile latency | ms | large_filtered_terms | 251.835 | 717.036 | 465.201 | 64.88% | 优 |
| 100th percentile latency | ms | large_filtered_terms | 260.27 | 749.665 | 489.395 | 65.28% | 优 |
| 50th percentile service time | ms | large_filtered_terms | 225.052 | 588.28 | 363.228 | 61.74% | 优 |
| 90th percentile service time | ms | large_filtered_terms | 233.16 | 595.415 | 362.255 | 60.84% | 优 |
| 99th percentile service time | ms | large_filtered_terms | 243.603 | 708.968 | 465.365 | 65.64% | 优 |
| 100th percentile service time | ms | large_filtered_terms | 252.129 | 741.922 | 489.793 | 66.02% | 优 |
| error rate | % | large_filtered_terms | 0 | 0 | 0 | - | - |
大量条件取反查询（query、must not） | Min Throughput | ops/s | large_prohibited_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | large_prohibited_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | large_prohibited_terms | 1.1 | 1.1 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | large_prohibited_terms | 235.179 | 610.41 | 375.231 | 61.47% | 优 |
| 90th percentile latency | ms | large_prohibited_terms | 241.076 | 631.544 | 390.468 | 61.83% | 优 |
| 99th percentile latency | ms | large_prohibited_terms | 255.983 | 774.92 | 518.937 | 66.97% | 优 |
| 100th percentile latency | ms | large_prohibited_terms | 259.046 | 779.168 | 520.122 | 66.75% | 优 |
| 50th percentile service time | ms | large_prohibited_terms | 227.487 | 603.342 | 375.855 | 62.30% | 优 |
| 90th percentile service time | ms | large_prohibited_terms | 233.792 | 624.55 | 390.758 | 62.57% | 优 |
| 99th percentile service time | ms | large_prohibited_terms | 248.53 | 767.511 | 518.981 | 67.62% | 优 |
| 100th percentile service time | ms | large_prohibited_terms | 251.083 | 771.655 | 520.572 | 67.46% | 优 |
| error rate | % | large_prohibited_terms | 0 | 0 | 0 | - | - |
降序排序查询 | Min Throughput | ops/s | desc_sort_population | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | desc_sort_population | 1.51 | 1.51 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | desc_sort_population | 1.51 | 1.51 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | desc_sort_population | 48.387 | 56.6622 | 8.2752 | 14.60% | 优 |
| 90th percentile latency | ms | desc_sort_population | 63.073 | 94.9756 | 31.9026 | 33.59% | 优 |
| 99th percentile latency | ms | desc_sort_population | 71.7498 | 97.2611 | 25.5113 | 26.23% | 优 |
| 100th percentile latency | ms | desc_sort_population | 83.3593 | 97.7817 | 14.4224 | 14.75% | 优 |
| 50th percentile service time | ms | desc_sort_population | 47.0436 | 55.1461 | 8.1025 | 14.69% | 优 |
| 90th percentile service time | ms | desc_sort_population | 61.3731 | 93.9102 | 32.5371 | 34.65% | 优 |
| 99th percentile service time | ms | desc_sort_population | 70.4811 | 95.6134 | 25.1323 | 26.29% | 优 |
| 100th percentile service time | ms | desc_sort_population | 81.6517 | 96.009 | 14.3573 | 14.95% | 优 |
| error rate | % | desc_sort_population | 0 | 0 | 0 | - | - |
升序排序查询 | Min Throughput | ops/s | asc_sort_population | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | asc_sort_population | 1.51 | 1.51 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | asc_sort_population | 1.51 | 1.51 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | asc_sort_population | 49.2469 | 57.9177 | 8.6708 | 14.97% | 优 |
| 90th percentile latency | ms | asc_sort_population | 67.5894 | 76.0044 | 8.415 | 11.07% | 优 |
| 99th percentile latency | ms | asc_sort_population | 84.6384 | 97.5143 | 12.8759 | 13.20% | 优 |
| 100th percentile latency | ms | asc_sort_population | 85.8124 | 97.5606 | 11.7482 | 12.04% | 优 |
| 50th percentile service time | ms | asc_sort_population | 47.8438 | 56.5861 | 8.7423 | 15.45% | 优 |
| 90th percentile service time | ms | asc_sort_population | 66.0821 | 74.7343 | 8.6522 | 11.58% | 优 |
| 99th percentile service time | ms | asc_sort_population | 83.6026 | 96.2868 | 12.6842 | 13.17% | 优 |
| 100th percentile service time | ms | asc_sort_population | 84.2175 | 96.6563 | 12.4388 | 12.87% | 优 |
| error rate | % | asc_sort_population | 0 | 0 | 0 | - | - |
升序排序后 after 跳转查询 | Min Throughput | ops/s | asc_sort_with_after_population | 1.5 | 1.5 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | asc_sort_with_after_population | 1.51 | 1.51 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | asc_sort_with_after_population | 1.51 | 1.51 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | asc_sort_with_after_population | 99.1943 | 83.423 | -15.7713 | -18.91% | 劣 |
| 90th percentile latency | ms | asc_sort_with_after_population | 86.0298 | 98.895 | 12.8652 | 13.01% | 优 |
| 99th percentile latency | ms | asc_sort_with_after_population | 102.268 | 131.329 | 29.061 | 22.13% | 优 |
| 100th percentile latency | ms | asc_sort_with_after_population | 106.33 | 132.25 | 25.92 | 19.60% | 优 |
| 50th percentile service time | ms | asc_sort_with_after_population | 68.2272 | 82.1201 | 13.8929 | 16.92% | 优 |
| 90th percentile service time | ms | asc_sort_with_after_population | 84.685 | 97.2085 | 12.5235 | 12.88% | 优 |
| 99th percentile service time | ms | asc_sort_with_after_population | 101.133 | 130.033 | 28.9 | 22.23% | 优 |
| 100th percentile service time | ms | asc_sort_with_after_population | 105.094 | 131.274 | 26.18 | 19.94% | 优 |
| 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 | ops/s | desc_sort_geonameid | 6.02 | 6.02 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | desc_sort_geonameid | 6.02 | 6.02 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | desc_sort_geonameid | 6.03 | 6.03 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | desc_sort_geonameid | 5.53008 | 5.15391 | -0.37617 | -7.30% | 劣 |
| 90th percentile latency | ms | desc_sort_geonameid | 6.20276 | 6.05507 | -0.14769 | -2.44% | 劣 |
| 99th percentile latency | ms | desc_sort_geonameid | 6.67673 | 7.41936 | 0.74263 | 10.01% | 优 |
| 100th percentile latency | ms | desc_sort_geonameid | 6.95103 | 24.1732 | 17.22217 | 71.24% | 优 |
| 50th percentile service time | ms | desc_sort_geonameid | 4.61231 | 4.38941 | -0.2229 | -5.08% | 劣 |
| 90th percentile service time | ms | desc_sort_geonameid | 5.45982 | 5.20354 | -0.25628 | -4.93% | 劣 |
| 99th percentile service time | ms | desc_sort_geonameid | 5.65304 | 5.75262 | 0.09958 | 1.73% | 优 |
| 100th percentile service time | ms | desc_sort_geonameid | 5.65578 | 23.3192 | 17.66342 | 75.75% | 优 |
| error rate | % | desc_sort_geonameid | 0 | 0 | 0 | - | - |
高基字段降序排序 after 跳转查询 | Min Throughput | ops/s | desc_sort_with_after_geonameid | 6.01 | 6.01 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | desc_sort_with_after_geonameid | 6.02 | 6.01 | -0.01 | -0.17% | 优 |
| Max Throughput | ops/s | desc_sort_with_after_geonameid | 6.02 | 6.02 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | desc_sort_with_after_geonameid | 56.5947 | 75.9327 | 19.338 | 25.47% | 优 |
| 90th percentile latency | ms | desc_sort_with_after_geonameid | 79.6503 | 88.6273 | 8.977 | 10.13% | 优 |
| 99th percentile latency | ms | desc_sort_with_after_geonameid | 87.7773 | 117.763 | 29.9857 | 25.46% | 优 |
| 100th percentile latency | ms | desc_sort_with_after_geonameid | 89.3947 | 118.003 | 28.6083 | 24.24% | 优 |
| 50th percentile service time | ms | desc_sort_with_after_geonameid | 55.4855 | 75.1241 | 19.6386 | 26.14% | 优 |
| 90th percentile service time | ms | desc_sort_with_after_geonameid | 79.2349 | 87.7269 | 8.492 | 9.68% | 优 |
| 99th percentile service time | ms | desc_sort_with_after_geonameid | 87.3803 | 116.924 | 29.5437 | 25.27% | 优 |
| 100th percentile service time | ms | desc_sort_with_after_geonameid | 88.3606 | 117.768 | 29.4074 | 24.97% | 优 |
| 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 | ops/s | asc_sort_geonameid | 6.02 | 6.02 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | asc_sort_geonameid | 6.02 | 6.02 | 0 | 0.00% | 优 |
| Max Throughput | ops/s | asc_sort_geonameid | 6.03 | 6.03 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | asc_sort_geonameid | 5.19317 | 4.49108 | -0.70209 | -15.63% | 劣 |
| 90th percentile latency | ms | asc_sort_geonameid | 5.74438 | 5.01747 | -0.72691 | -14.49% | 劣 |
| 99th percentile latency | ms | asc_sort_geonameid | 6.22846 | 5.49664 | -0.73182 | -13.31% | 劣 |
| 100th percentile latency | ms | asc_sort_geonameid | 11.6377 | 5.53539 | -6.10231 | -110.24% | 劣 |
| 50th percentile service time | ms | asc_sort_geonameid | 4.35586 | 3.56597 | -0.78989 | -22.15% | 劣 |
| 90th percentile service time | ms | asc_sort_geonameid | 4.92152 | 3.97107 | -0.95045 | -23.93% | 劣 |
| 99th percentile service time | ms | asc_sort_geonameid | 5.38949 | 4.33548 | -1.05401 | -24.31% | 劣 |
| 100th percentile service time | ms | asc_sort_geonameid | 10.6436 | 4.5666 | -6.077 | -133.07% | 劣 |
| error rate | % | asc_sort_geonameid | 0 | 0 | 0 | - | - |
高基字段升序排序 after 跳转查询 | Min Throughput | ops/s | asc_sort_with_after_geonameid | 6.01 | 6.01 | 0 | 0.00% | 优 |
| Median Throughput | ops/s | asc_sort_with_after_geonameid | 6.02 | 6.01 | -0.01 | -0.17% | 优 |
| Max Throughput | ops/s | asc_sort_with_after_geonameid | 6.02 | 6.02 | 0 | 0.00% | 优 |
| 50th percentile latency | ms | asc_sort_with_after_geonameid | 58.1403 | 69.5377 | 11.3974 | 16.39% | 优 |
| 90th percentile latency | ms | asc_sort_with_after_geonameid | 76.5695 | 81.7929 | 5.2234 | 6.39% | 优 |
| 99th percentile latency | ms | asc_sort_with_after_geonameid | 91.6296 | 98.1089 | 6.4793 | 6.60% | 优 |
| 100th percentile latency | ms | asc_sort_with_after_geonameid | 91.6364 | 104.363 | 12.7266 | 12.19% | 优 |
| 50th percentile service time | ms | asc_sort_with_after_geonameid | 57.1683 | 68.532 | 11.3637 | 16.58% | 优 |
| 90th percentile service time | ms | asc_sort_with_after_geonameid | 75.7573 | 81.0123 | 5.255 | 6.49% | 优 |
| 99th percentile service time | ms | asc_sort_with_after_geonameid | 91.1533 | 97.807 | 6.6537 | 6.80% | 优 |
| 100th percentile service time | ms | asc_sort_with_after_geonameid | 91.3662 | 103.404 | 12.0378 | 11.64% | 优 |
| error rate | % | asc_sort_with_after_geonameid | 0 | 0 | 0 | - | - |
