This topic lists the performance metrics of an Elasticsearch cluster that contains three data nodes. Each data node has 2 vCPUs and 8 GiB of memory. The metrics include the Kibana metrics during the performance test and the performance metrics that are used to calculate these Kibana metrics.
Note The official GeoNames database is used for the test. The database contains 11,520,617 DOC files whose total
size is 3.3 GB.
Kibana metrics during the performance test
Performance metrics
Note For more information about the metrics and operations listed in the following table,
see Overview.
| Metric | Operation | Value | Unit |
|---|---|---|---|
| Indexing time | None | 23.9479 | min |
| Merge time | None | 14.3001 | min |
| Refresh time | None | 5.26405 | min |
| Flush time | None | 0.0308333 | min |
| Merge throttle time | None | 1.27945 | min |
| Total Young Gen GC | None | 183.74 | s |
| Total Old Gen GC | None | 1.125 | s |
| Heap used for segments | None | 18.8167 | MB |
| Heap used for doc values | None | 0.452751 | MB |
| Heap used for terms | None | 17.2004 | MB |
| Heap used for norms | None | 0.0852051 | MB |
| Heap used for points | None | 0.241465 | MB |
| Heap used for stored fields | None | 0.836876 | MB |
| Segment count | None | 140 | items |
| Min Throughput | index-append | 28115.4 | docs/s |
| Median Throughput | index-append | 28645.5 | docs/s |
| Max Throughput | index-append | 30037.8 | docs/s |
| 50th percentile latency | index-append | 1447.76 | ms |
| 90th percentile latency | index-append | 1847.05 | ms |
| 99th percentile latency | index-append | 2264.68 | ms |
| 99.9th percentile latency | index-append | 2515.95 | ms |
| 100th percentile latency | index-append | 2608.68 | ms |
| 50th percentile service time | index-append | 1447.76 | ms |
| 90th percentile service time | index-append | 1847.05 | ms |
| 99th percentile service time | index-append | 2264.68 | ms |
| 99.9th percentile service time | index-append | 2515.95 | ms |
| 100th percentile service time | index-append | 2608.68 | ms |
| error rate | index-append | 0 | % |
| Min Throughput | force-merge | 2.1 | ops/s |
| Median Throughput | force-merge | 2.1 | ops/s |
| Max Throughput | force-merge | 2.1 | ops/s |
| 100th percentile latency | force-merge | 475.984 | ms |
| 100th percentile service time | force-merge | 475.984 | ms |
| error rate | force-merge | 0 | % |
| Min Throughput | index-stats | 97.75 | ops/s |
| Median Throughput | index-stats | 100.05 | ops/s |
| Max Throughput | index-stats | 100.07 | ops/s |
| 50th percentile latency | index-stats | 5.09015 | ms |
| 90th percentile latency | index-stats | 10.7365 | ms |
| 99th percentile latency | index-stats | 234.761 | ms |
| 99.9th percentile latency | index-stats | 277.393 | ms |
| 100th percentile latency | index-stats | 281.866 | ms |
| 50th percentile service time | index-stats | 5.01096 | ms |
| 90th percentile service time | index-stats | 5.30021 | ms |
| 99th percentile service time | index-stats | 12.0005 | ms |
| 99.9th percentile service time | index-stats | 141.631 | ms |
| 100th percentile service time | index-stats | 150.153 | ms |
| error rate | index-stats | 0 | % |
| Min Throughput | node-stats | 100.01 | ops/s |
| Median Throughput | node-stats | 100.08 | ops/s |
| Max Throughput | node-stats | 100.49 | ops/s |
| 50th percentile latency | node-stats | 4.90659 | ms |
| 90th percentile latency | node-stats | 5.29285 | ms |
| 99th percentile latency | node-stats | 29.3245 | ms |
| 99.9th percentile latency | node-stats | 43.3885 | ms |
| 100th percentile latency | node-stats | 44.6019 | ms |
| 50th percentile service time | node-stats | 4.83552 | ms |
| 90th percentile service time | node-stats | 5.12694 | ms |
| 99th percentile service time | node-stats | 9.08739 | ms |
| 99.9th percentile service time | node-stats | 39.744 | ms |
| 100th percentile service time | node-stats | 44.5383 | ms |
| error rate | node-stats | 0 | % |
| Min Throughput | default | 47.83 | ops/s |
| Median Throughput | default | 48.28 | ops/s |
| Max Throughput | default | 48.73 | ops/s |
| 50th percentile latency | default | 617.465 | ms |
| 90th percentile latency | default | 1033.98 | ms |
| 99th percentile latency | default | 1083.23 | ms |
| 99.9th percentile latency | default | 1095.4 | ms |
| 100th percentile latency | default | 1097.14 | ms |
| 50th percentile service time | default | 18.646 | ms |
| 90th percentile service time | default | 24.9381 | ms |
| 99th percentile service time | default | 35.7667 | ms |
| 99.9th percentile service time | default | 57.3679 | ms |
| 100th percentile service time | default | 151.505 | ms |
| error rate | default | 0 | % |
| Min Throughput | term | 199.43 | ops/s |
| Median Throughput | term | 200.07 | ops/s |
| Max Throughput | term | 200.13 | ops/s |
| 50th percentile latency | term | 2.9728 | ms |
| 90th percentile latency | term | 7.10648 | ms |
| 99th percentile latency | term | 22.4487 | ms |
| 99.9th percentile latency | term | 29.0737 | ms |
| 100th percentile latency | term | 29.6253 | ms |
| 50th percentile service time | term | 2.87833 | ms |
| 90th percentile service time | term | 3.08983 | ms |
| 99th percentile service time | term | 19.9777 | ms |
| 99.9th percentile service time | term | 29.0082 | ms |
| 100th percentile service time | term | 29.5597 | ms |
| error rate | term | 0 | % |
| Min Throughput | phrase | 199.71 | ops/s |
| Median Throughput | phrase | 200.04 | ops/s |
| Max Throughput | phrase | 200.07 | ops/s |
| 50th percentile latency | phrase | 3.61484 | ms |
| 90th percentile latency | phrase | 16.5523 | ms |
| 99th percentile latency | phrase | 31.394 | ms |
| 99.9th percentile latency | phrase | 33.902 | ms |
| 100th percentile latency | phrase | 34.5784 | ms |
| 50th percentile service time | phrase | 3.47402 | ms |
| 90th percentile service time | phrase | 3.90958 | ms |
| 99th percentile service time | phrase | 19.3773 | ms |
| 99.9th percentile service time | phrase | 22.7947 | ms |
| 100th percentile service time | phrase | 27.8164 | ms |
| error rate | phrase | 0 | % |
| Min Throughput | country_agg_uncached | 4.63 | ops/s |
| Median Throughput | country_agg_uncached | 4.65 | ops/s |
| Max Throughput | country_agg_uncached | 4.67 | ops/s |
| 50th percentile latency | country_agg_uncached | 14864.3 | ms |
| 90th percentile latency | country_agg_uncached | 21046 | ms |
| 99th percentile latency | country_agg_uncached | 22902 | ms |
| 99.9th percentile latency | country_agg_uncached | 22997.6 | ms |
| 100th percentile latency | country_agg_uncached | 23018.7 | ms |
| 50th percentile service time | country_agg_uncached | 204.174 | ms |
| 90th percentile service time | country_agg_uncached | 242.492 | ms |
| 99th percentile service time | country_agg_uncached | 345.382 | ms |
| 99.9th percentile service time | country_agg_uncached | 378.302 | ms |
| 100th percentile service time | country_agg_uncached | 422.53 | ms |
| error rate | country_agg_uncached | 0 | % |
| Min Throughput | country_agg_cached | 98.37 | ops/s |
| Median Throughput | country_agg_cached | 100.06 | ops/s |
| Max Throughput | country_agg_cached | 100.13 | ops/s |
| 50th percentile latency | country_agg_cached | 3.2638 | ms |
| 90th percentile latency | country_agg_cached | 4.69259 | ms |
| 99th percentile latency | country_agg_cached | 189.143 | ms |
| 99.9th percentile latency | country_agg_cached | 249.851 | ms |
| 100th percentile latency | country_agg_cached | 256.028 | ms |
| 50th percentile service time | country_agg_cached | 3.18679 | ms |
| 90th percentile service time | country_agg_cached | 3.42086 | ms |
| 99th percentile service time | country_agg_cached | 20.4171 | ms |
| 99.9th percentile service time | country_agg_cached | 117.273 | ms |
| 100th percentile service time | country_agg_cached | 255.951 | ms |
| error rate | country_agg_cached | 0 | % |
| Min Throughput | scroll | 59.16 | ops/s |
| Median Throughput | scroll | 60.44 | ops/s |
| Max Throughput | scroll | 61.02 | ops/s |
| 50th percentile latency | scroll | 168347 | ms |
| 90th percentile latency | scroll | 240658 | ms |
| 99th percentile latency | scroll | 257048 | ms |
| 100th percentile latency | scroll | 258853 | ms |
| 50th percentile service time | scroll | 402.962 | ms |
| 90th percentile service time | scroll | 431.267 | ms |
| 99th percentile service time | scroll | 455.632 | ms |
| 100th percentile service time | scroll | 601.214 | ms |
| error rate | scroll | 0 | % |
| Min Throughput | expression | 2 | ops/s |
| Median Throughput | expression | 2 | ops/s |
| Max Throughput | expression | 2 | ops/s |
| 50th percentile latency | expression | 409.417 | ms |
| 90th percentile latency | expression | 434.858 | ms |
| 99th percentile latency | expression | 501.498 | ms |
| 100th percentile latency | expression | 517.438 | ms |
| 50th percentile service time | expression | 409.165 | ms |
| 90th percentile service time | expression | 434.749 | ms |
| 99th percentile service time | expression | 498.681 | ms |
| 100th percentile service time | expression | 517.332 | ms |
| error rate | expression | 0 | % |
| Min Throughput | painless_static | 1.96 | ops/s |
| Median Throughput | painless_static | 1.97 | ops/s |
| Max Throughput | painless_static | 1.97 | ops/s |
| 50th percentile latency | painless_static | 3163.94 | ms |
| 90th percentile latency | painless_static | 3679.27 | ms |
| 99th percentile latency | painless_static | 3994.52 | ms |
| 100th percentile latency | painless_static | 4006.5 | ms |
| 50th percentile service time | painless_static | 503.588 | ms |
| 90th percentile service time | painless_static | 528.807 | ms |
| 99th percentile service time | painless_static | 600.103 | ms |
| 100th percentile service time | painless_static | 623.666 | ms |
| error rate | painless_static | 0 | % |
| Min Throughput | painless_dynamic | 2 | ops/s |
| Median Throughput | painless_dynamic | 2 | ops/s |
| Max Throughput | painless_dynamic | 2 | ops/s |
| 50th percentile latency | painless_dynamic | 611.305 | ms |
| 90th percentile latency | painless_dynamic | 786.806 | ms |
| 99th percentile latency | painless_dynamic | 973.432 | ms |
| 100th percentile latency | painless_dynamic | 982.484 | ms |
| 50th percentile service time | painless_dynamic | 494.097 | ms |
| 90th percentile service time | painless_dynamic | 518.082 | ms |
| 99th percentile service time | painless_dynamic | 606.748 | ms |
| 100th percentile service time | painless_dynamic | 638.903 | ms |
| error rate | painless_dynamic | 0 | % |
| Min Throughput | large_filtered_terms | 1.39 | ops/s |
| Median Throughput | large_filtered_terms | 1.4 | ops/s |
| Max Throughput | large_filtered_terms | 1.4 | ops/s |
| 50th percentile latency | large_filtered_terms | 65601.1 | ms |
| 90th percentile latency | large_filtered_terms | 82494.7 | ms |
| 99th percentile latency | large_filtered_terms | 86452.2 | ms |
| 100th percentile latency | large_filtered_terms | 86857.3 | ms |
| 50th percentile service time | large_filtered_terms | 707.17 | ms |
| 90th percentile service time | large_filtered_terms | 747.949 | ms |
| 99th percentile service time | large_filtered_terms | 847.069 | ms |
| 100th percentile service time | large_filtered_terms | 927.917 | ms |
| error rate | large_filtered_terms | 0 | % |
| Min Throughput | large_prohibited_terms | 1.46 | ops/s |
| Median Throughput | large_prohibited_terms | 1.46 | ops/s |
| Max Throughput | large_prohibited_terms | 1.46 | ops/s |
| 50th percentile latency | large_prohibited_terms | 55916.3 | ms |
| 90th percentile latency | large_prohibited_terms | 70529.7 | ms |
| 99th percentile latency | large_prohibited_terms | 73769.1 | ms |
| 100th percentile latency | large_prohibited_terms | 74143.9 | ms |
| 50th percentile service time | large_prohibited_terms | 679.394 | ms |
| 90th percentile service time | large_prohibited_terms | 717.476 | ms |
| 99th percentile service time | large_prohibited_terms | 782.085 | ms |
| 100th percentile service time | large_prohibited_terms | 822.723 | ms |
| error rate | large_prohibited_terms | 0 | % |