This topic describes the performance test results of a general-purpose ApsaraDB RDS instance that runs MySQL 5.6.

Note The performance test results are for reference only. For more information about how to use an ApsaraDB RDS instance that runs MySQL 5.6, see Troubleshoot slow SQL statements on an ApsaraDB RDS for MySQL instance.

Scenario 1: Cache-based queries

All data can be read to the cache for queries.

Set the table_size parameter to 25000 and the tables parameter to 250 in SysBench.

Scenario 1 based on ApsaraDB RDS for MySQL 5.6
CPU

(number of cores)

Memory capacity

(GB)

Maximum number of connections IOPS Number of SysBench threads SysBench QPS SysBench TPS
1 1 300 600 8 638.25 31.91
1 2 600 1,000 8 1,774.43 88.72
2 4 1,200 2,000 8 18,561.51 928.08
2 8 2,000 4,000 8 18,672.65 933.63
4 8 2,000 5,000 16 47,272.72 2,363.64
4 16 4,000 7,000 16 48,104.96 2,405.25
8 16 4,000 8,000 32 71,899.12 3,594.96
8 32 8,000 12,000 32 73,847.17 3,692.36
16 64 16,000 14,000 64 107,670.53 5,383.53
16 96 24,000 16,000 64 108,473.73 5,423.69

Scenario 2: Disk I/O-based queries

Only part of the data is read to the cache. The RDS instance needs to read and write data to the disk and update the cache to complete queries.

Set the table_size parameter to 800000 and the tables parameter to 150 in SysBench.

Scenario 2 based on ApsaraDB RDS for MySQL 5.6
CPU

(number of cores)

Memory capacity

(GB)

Maximum number of connections IOPS Number of SysBench threads SysBench QPS SysBench TPS
1 1 300 600 64 395.41 19.77
1 2 600 1,000 64 682.55 34.13
2 4 1,200 2,000 64 1,477.06 73.85
2 8 2,000 4,000 64 3,395.79 169.79
4 8 2,000 5,000 64 4,245.29 212.26
4 16 4,000 7,000 64 7,759.08 387.95
8 16 4,000 8,000 64 8,865.62 443.28
8 32 8,000 12,000 64 28,084.63 1,404.23
16 64 16,000 14,000 64 97,350.85 4,867.54
16 96 24,000 16,000 64 98,914.56 4,945.73