Test Data of Master/Standby DCS Redis 4.0 or 5.0 Instances
Test Environment
Test Command
redis-benchmark -h {IP} -p {Port} -a {password} -n {nreqs} -r {randomkeys} -c {connection} -d {datasize} -t {command} Reference values: -c {connect_number}: 500; -n {nreqs}: 10000000; -r {randomkeys}: 1000000; -d {datasize}: 32; -t {command}: set
For details about the test method and parameters, see Testing Redis Performance Using redis-benchmark.
Test Result
- The following test results are for reference only. The performance may vary depending on the site environment and network fluctuation.
- Certain cache sizes are taken for example in the following test results. For more information, see DCS Instance Specifications.
- QPS: Query per second, indicates number of read and write operations per second. Unit: count/second.
- Average Latency: Average latency of operations, in milliseconds.
- xth Percentile Latency: latency of x% of operations, in milliseconds. For example, if the value is 10 ms, 99.99th percentile latency indicates that 99.99% queries can be processed within 10 ms.
| Redis Cache Size | CPU Type | Concurrent Connections | QPS | 99.99th-Percentile Latency (ms) | First 100th-Percentile Latency (ms) | Last 100th-Percentile Latency (ms) | Average Latency (ms) |
|---|---|---|---|---|---|---|---|
| 8 GB | x86 | 500 | 132,068.98 | 11 | 18 | 205 | 3.298 |
| 10,000 | 82,386.58 | 171 | 178 | 263 | 69.275 | ||
| 32 GB | x86 | 500 | 131,385.33 | 9.5 | 16 | 17 | 3.333 |
| 10,000 | 82,275.41 | 157 | 162.18 | 162.43 | 62.105 |
| Redis Cache Size | CPU Type | Concurrent Connections | QPS | 99.99th-Percentile Latency (ms) | First 100th-Percentile Latency (ms) | Last 100th-Percentile Latency (ms) | Average Latency (ms) |
|---|---|---|---|---|---|---|---|
| 8 GB | x86 | 500 | 138,652.02 | 7 | 11 | 12 | 2.117 |
| 10,000 | 82,710.94 | 123.7 | 281.6 | 282.9 | 61.078 | ||
| 32 GB | x86 | 500 | 139,113.02 | 6.6 | 10 | 11 | 2.119 |
| 10,000 | 82,489.36 | 100 | 105.66 | 106 | 60.968 |
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.