Last updated: 2020-05-09 17:44:58
Yahoo! Cloud Serving Benchmark.
Testing Method
- A single data entry is 1 KB in size.
- Import up to 80% of the instance capacity. For example, if the instance capacity is 100 GB, import 80 GB of data into the instance.
- Perform 50% read and 50% update operations to get the QPS (queries per second) data.
- As performance varies by scenario, please select instances with appropriate specifications according to your business requirements and the test data.
| CPU |
MEM |
QPS |
| 1 core |
2 GB |
3,000 |
| 2 cores |
4 GB |
5,000 |
| 2 cores |
6 GB |
6,000 |
| 4 cores |
8 GB |
9,000 |
| 4 cores |
12 GB |
14,000 |
| 6 cores |
16 GB |
20,000 |
| 10 cores |
24 GB |
25,000 |
| 12 cores |
32 GB |
27,000 |
| 18 cores |
48 GB |
30,000 |
| 24 cores |
64 GB |
33,000 |
| CPU |
MEM |
Number of Collections |
Number of Documents per Collection |
Test Dataset (GB) |
Runtime (s) |
Data Sampling Interval (s) |
Average QPS (Rounded) |
| 2 cores |
4 GB |
2 |
100 million |
79 |
1,800 |
10 |
5,000 |
| 4 cores |
8 GB |
3 |
100 million |
118 |
1,800 |
10 |
9,000 |
| 6 cores |
16 GB |
6 |
100 million |
237 |
1,800 |
10 |
20,000 |
| 12 cores |
32 GB |
6 |
200 million |
426 |
1,800 |
10 |
27,000 |
| 24 cores |
64 GB |
15 |
200 million |
1,161 |
1,800 |
10 |
33,000 |
| 24 cores |
128 GB |
30 |
200 million |
2,317 |
1,800 |
10 |
36,000 |
| 32 cores |
240 GB |
40 |
200 million |
3,031 |
1,800 |
10 |
39,000 |
Was this page helpful?