Introduction

This is a report for the insert benchmark with 20M docs and 4 client(s). It is generated by scripts (bash, awk, sed) and Tufte might not be impressed. An overview of the insert benchmark is here and a short update is here. Below, by DBMS, I mean DBMS+version.config. An example is my8020.c10b40 where my means MySQL, 8020 is version 8.0.20 and c10b40 is the name for the configuration file.

The test server has 8 AMD cores, 16G RAM and an NVMe SSD. It is described here as the Beelink. The benchmark was run with 4 clients and there were 1 or 2 connections per client (1 for queries, 1 for inserts). The benchmark loads 20M rows without secondary indexes, creates secondary indexes, loads another 20M rows then does 3 read+write tests for one hour each that do queries as fast as possible with 100, 500 and then 1000 writes/second/client concurrent with the queries. There is one table per client. The database is cached by the OS page cache but not by the storage engine. Clients and the DBMS share one server. The per-database configs are in the per-database subdirectories here.

The tested DBMS are:


Contents


Summary

The numbers are inserts/s for l.i0 and l.i1, indexed docs (or rows) /s for l.x and queries/s for q*.2. The values are the average rate over the entire test for inserts (IPS) and queries (QPS). The range of values for IPS and QPS is split into 3 parts: bottom 25%, middle 50%, top 25%. Values in the bottom 25% have a red background, values in the top 25% have a green background and values in the middle have no color. A gray background is used for values that can be ignored because the DBMS did not sustain the target insert rate. Red backgrounds are not used when the minimum value is within 80% of the max value.

dbmsl.i0l.xl.i1q100.1q500.1q1000.1
20m.fbmy5635_rel.cay9c1g 19802029558892592105141026910270
20m.fbmy8028_rel_native_lto.cay9c1g 1600002871438333310244100679909

This table has relative throughput, throughput for the DBMS relative to the DBMS in the first line, using the absolute throughput from the previous table.

dbmsl.i0l.xl.i1q100.1q500.1q1000.1
20m.fbmy5635_rel.cay9c1g 1.001.001.001.001.001.00
20m.fbmy8028_rel_native_lto.cay9c1g 0.810.970.900.970.980.96

This lists the average rate of inserts/s for the tests that do inserts concurrent with queries. For such tests the query rate is listed in the table above. The read+write tests are setup so that the insert rate should match the target rate every second. Cells that are not at least 95% of the target have a red background to indicate a failure to satisfy the target.

dbmsq100.1q500.1q1000.1
fbmy5635_rel.cay9c1g39919943989
fbmy8028_rel_native_lto.cay9c1g39919943989
target40020004000

l.i0

l.i0: load without secondary indexes. Graphs for performance per 1-second interval are here.

Average throughput:

Image

Insert response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For the max column values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g99.7490.2370.0010.0130.107
fbmy8028_rel_native_lto.cay9c1g99.3020.4830.2020.0130.139

Performance metrics for the DBMS listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
198020	0	0	0.0	122.2	51.5	0.000	0.000	0.001	0.267	20134	72.1	0.102	29	0.9	1.9	0.4	0.107	55157	22284	20m.fbmy5635_rel.cay9c1g
160000	0	0	0.0	111.2	44.3	0.000	0.000	0.001	0.284	16148	66.6	0.101	33	0.7	1.9	0.8	0.139	43253	29267	20m.fbmy8028_rel_native_lto.cay9c1g

l.x

l.x: create secondary indexes.

Average throughput:

Image

Performance metrics for the DBMS listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
295588	0	1	0.0	131.8	55.7	0.000	0.000	0.000	0.193	656	48.6	0.002	13	1.4	2.4	1.6	0.003	NA	NA	20m.fbmy5635_rel.cay9c1g
287143	0	1	0.0	113.8	47.4	0.000	0.000	0.000	0.169	813	47.3	0.003	13	1.4	2.7	2.2	0.008	NA	NA	20m.fbmy8028_rel_native_lto.cay9c1g

l.i1

l.i1: continue load after secondary indexes created. Graphs for performance per 1-second interval are here.

Average throughput:

Image

Insert response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For the max column values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g99.6110.3790.0040.0070.108
fbmy8028_rel_native_lto.cay9c1g99.2500.6370.1060.0070.108

Performance metrics for the DBMS listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
92592	0	12	1.5	173.4	69.9	0.000	0.017	0.002	0.773	20258	73.9	0.219	64	4.3	6.0	1.3	0.108	23931	17836	20m.fbmy5635_rel.cay9c1g
83333	0	16	1.9	168.4	65.4	0.000	0.023	0.002	0.804	17939	74.3	0.215	71	3.5	5.4	1.7	0.108	21377	15783	20m.fbmy8028_rel_native_lto.cay9c1g

q100.1

q100.1: range queries with 100 insert/s per client. Graphs for performance per 1-second interval are here.

Average throughput:

Image

Query response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For max values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g10.29989.6930.0070.0010.012
fbmy8028_rel_native_lto.cay9c1g7.17192.8180.010nonzero0.014

Insert response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For max values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g99.4930.5070.011
fbmy8028_rel_native_lto.cay9c1g99.6530.3190.0280.017

Performance metrics for the DBMS listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
399	10514	17	0.2	7.9	1.9	0.002	0.023	0.020	4.992	39547	50.6	3.762	385	2.8	4.7	1.4	0.012	2637	2253	20m.fbmy5635_rel.cay9c1g
399	10244	14	0.2	7.4	1.7	0.001	0.015	0.019	4.458	38719	51.2	3.780	400	2.8	4.8	1.8	0.014	2573	2047	20m.fbmy8028_rel_native_lto.cay9c1g

q500.1

q500.1: range queries with 500 insert/s per client. Graphs for performance per 1-second interval are here.

Average throughput:

Image

Query response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For max values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g7.75892.2110.0300.001nonzero0.021
fbmy8028_rel_native_lto.cay9c1g4.12595.8260.0460.002nonzero0.020

Insert response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For max values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g97.8932.0900.0170.046
fbmy8028_rel_native_lto.cay9c1g97.8692.0460.0830.0010.065

Performance metrics for the DBMS listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
1994	10269	0	0.0	7.8	2.0	0.000	0.000	0.004	1.014	38467	51.1	3.746	398	3.2	3.5	1.4	0.021	2557	2238	20m.fbmy5635_rel.cay9c1g
1994	10067	0	0.0	7.1	1.6	0.000	0.000	0.004	0.845	38005	51.6	3.775	410	3.2	3.7	1.7	0.020	2493	2062	20m.fbmy8028_rel_native_lto.cay9c1g

q1000.1

q1000.1: range queries with 1000 insert/s per client. Graphs for performance per 1-second interval are here.

Average throughput:

Image

Query response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For max values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g6.20393.7160.0770.004nonzero0.018
fbmy8028_rel_native_lto.cay9c1g2.86697.0160.1130.005nonzero0.019

Insert response time histogram: each cell has the percentage of responses that take <= the time in the header and max is the max response time in seconds. For max values in the top 25% of the range have a red background and in the bottom 25% of the range have a green background. The red background is not used when the min value is within 80% of the max value.

dbms256us1ms4ms16ms64ms256ms1s4s16sgtmax
fbmy5635_rel.cay9c1g93.8086.1510.0420.056
fbmy8028_rel_native_lto.cay9c1g94.7065.1660.1280.061

Performance metrics for the DBMS listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
3989	10270	0	0.0	15.4	5.0	0.000	0.000	0.004	1.290	37504	53.3	3.652	415	3.7	4.9	1.4	0.018	2509	2207	20m.fbmy5635_rel.cay9c1g
3989	9909	0	0.0	15.3	4.9	0.000	0.000	0.004	1.262	36735	53.8	3.707	434	3.7	5.1	1.8	0.019	2493	2014	20m.fbmy8028_rel_native_lto.cay9c1g

l.i0

l.i0: load without secondary indexes

Performance metrics for all DBMS, not just the ones listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
198020	0	0	0.0	122.2	51.5	0.000	0.000	0.001	0.267	20134	72.1	0.102	29	0.9	1.9	0.4	0.107	55157	22284	20m.fbmy5635_rel.cay9c1g
160000	0	0	0.0	111.2	44.3	0.000	0.000	0.001	0.284	16148	66.6	0.101	33	0.7	1.9	0.8	0.139	43253	29267	20m.fbmy8028_rel_native_lto.cay9c1g

l.x

l.x: create secondary indexes

Performance metrics for all DBMS, not just the ones listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
295588	0	1	0.0	131.8	55.7	0.000	0.000	0.000	0.193	656	48.6	0.002	13	1.4	2.4	1.6	0.003	NA	NA	20m.fbmy5635_rel.cay9c1g
287143	0	1	0.0	113.8	47.4	0.000	0.000	0.000	0.169	813	47.3	0.003	13	1.4	2.7	2.2	0.008	NA	NA	20m.fbmy8028_rel_native_lto.cay9c1g

l.i1

l.i1: continue load after secondary indexes created

Performance metrics for all DBMS, not just the ones listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
92592	0	12	1.5	173.4	69.9	0.000	0.017	0.002	0.773	20258	73.9	0.219	64	4.3	6.0	1.3	0.108	23931	17836	20m.fbmy5635_rel.cay9c1g
83333	0	16	1.9	168.4	65.4	0.000	0.023	0.002	0.804	17939	74.3	0.215	71	3.5	5.4	1.7	0.108	21377	15783	20m.fbmy8028_rel_native_lto.cay9c1g

q100.1

q100.1: range queries with 100 insert/s per client

Performance metrics for all DBMS, not just the ones listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
399	10514	17	0.2	7.9	1.9	0.002	0.023	0.020	4.992	39547	50.6	3.762	385	2.8	4.7	1.4	0.012	2637	2253	20m.fbmy5635_rel.cay9c1g
399	10244	14	0.2	7.4	1.7	0.001	0.015	0.019	4.458	38719	51.2	3.780	400	2.8	4.8	1.8	0.014	2573	2047	20m.fbmy8028_rel_native_lto.cay9c1g

q500.1

q500.1: range queries with 500 insert/s per client

Performance metrics for all DBMS, not just the ones listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
1994	10269	0	0.0	7.8	2.0	0.000	0.000	0.004	1.014	38467	51.1	3.746	398	3.2	3.5	1.4	0.021	2557	2238	20m.fbmy5635_rel.cay9c1g
1994	10067	0	0.0	7.1	1.6	0.000	0.000	0.004	0.845	38005	51.6	3.775	410	3.2	3.7	1.7	0.020	2493	2062	20m.fbmy8028_rel_native_lto.cay9c1g

q1000.1

q1000.1: range queries with 1000 insert/s per client

Performance metrics for all DBMS, not just the ones listed above. Some are normalized by throughput, others are not. Legend for results is here.

ips	qps	rps	rmbps	wps	wmbps	rpq	rkbpq	wpi	wkbpi	csps	cpups	cspq	cpupq	dbgb1	dbgb2	rss	maxop	p50	p99	tag
3989	10270	0	0.0	15.4	5.0	0.000	0.000	0.004	1.290	37504	53.3	3.652	415	3.7	4.9	1.4	0.018	2509	2207	20m.fbmy5635_rel.cay9c1g
3989	9909	0	0.0	15.3	4.9	0.000	0.000	0.004	1.262	36735	53.8	3.707	434	3.7	5.1	1.8	0.019	2493	2014	20m.fbmy8028_rel_native_lto.cay9c1g

l.i0

Insert response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
0.000	0.000	99.749	0.237	0.001	0.013	0.000	0.000	0.000	0.000	0.107	fbmy5635_rel.cay9c1g
0.000	0.000	99.302	0.483	0.202	0.013	0.000	0.000	0.000	0.000	0.139	fbmy8028_rel_native_lto.cay9c1g

l.x

TODO - determine whether there is data for create index response time


l.i1

Insert response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
0.000	0.000	99.611	0.379	0.004	0.007	0.000	0.000	0.000	0.000	0.108	fbmy5635_rel.cay9c1g
0.000	0.000	99.250	0.637	0.106	0.007	0.000	0.000	0.000	0.000	0.108	fbmy8028_rel_native_lto.cay9c1g

q100.1

Query response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
10.299	89.693	0.007	0.001	0.000	0.000	0.000	0.000	0.000	0.000	0.012	fbmy5635_rel.cay9c1g
7.171	92.818	0.010	nonzero	0.000	0.000	0.000	0.000	0.000	0.000	0.014	fbmy8028_rel_native_lto.cay9c1g

Insert response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
0.000	0.000	99.493	0.507	0.000	0.000	0.000	0.000	0.000	0.000	0.011	fbmy5635_rel.cay9c1g
0.000	0.000	99.653	0.319	0.028	0.000	0.000	0.000	0.000	0.000	0.017	fbmy8028_rel_native_lto.cay9c1g

q500.1

Query response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
7.758	92.211	0.030	0.001	nonzero	0.000	0.000	0.000	0.000	0.000	0.021	fbmy5635_rel.cay9c1g
4.125	95.826	0.046	0.002	nonzero	0.000	0.000	0.000	0.000	0.000	0.020	fbmy8028_rel_native_lto.cay9c1g

Insert response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
0.000	0.000	97.893	2.090	0.017	0.000	0.000	0.000	0.000	0.000	0.046	fbmy5635_rel.cay9c1g
0.000	0.000	97.869	2.046	0.083	0.001	0.000	0.000	0.000	0.000	0.065	fbmy8028_rel_native_lto.cay9c1g

q1000.1

Query response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
6.203	93.716	0.077	0.004	nonzero	0.000	0.000	0.000	0.000	0.000	0.018	fbmy5635_rel.cay9c1g
2.866	97.016	0.113	0.005	nonzero	0.000	0.000	0.000	0.000	0.000	0.019	fbmy8028_rel_native_lto.cay9c1g

Insert response time histogram

256us	1ms	4ms	16ms	64ms	256ms	1s	4s	16s	gt	max	tag
0.000	0.000	93.808	6.151	0.042	0.000	0.000	0.000	0.000	0.000	0.056	fbmy5635_rel.cay9c1g
0.000	0.000	94.706	5.166	0.128	0.000	0.000	0.000	0.000	0.000	0.061	fbmy8028_rel_native_lto.cay9c1g