This topic describes the test environment, test plan, and benchmark results for the TPC-H test on OceanBase Database.
Test environment
The example below uses Alibaba Cloud Elastic Compute Service (ECS).
Hardware configuration
Service typeECS typeNumber of instancesNumber of CPU coresMemoryOceanBase Database ecs.r8i.8xlarge 3 32C 256G
Each server has a 400 GB log disk and a 1,200 GB data disk, and the performance level is PL1.Software version
Service typeSoftware versionOceanBase Database - Enterprise Edition: OceanBase 4.4.2.0
- Community Edition: OceanBase_CE 4.4.2.0
TPC-H V3.0.0 OS CentOS Linux release 7.9.2009 (Core) Tenant specifications
obd cluster tenant create obperf -n tpcc_tenant --max-cpu=28 --memory-size=180G --zone-list=zone1,zone2,zone3 --primary-zone=RANDOM --locality=F@zone1,F@zone2,F@zone3 --charset=utf8 -s 'ob_tcp_invited_nodes="%"' --optimize=htapNote
In this example,
--optimize=htapis the default load type. In a production environment, select an appropriate load type based on the cluster type.
Test plan
Deploy an OceanBase cluster with OBD. Run the TPC-H client on one server as the load generator. ODP is not required—you can connect directly to any OBServer node.
Use a three-node OceanBase cluster with a 1:1:1 deployment ratio. After deployment, create a tenant and user for TPC-H testing. Do not use the
systenant; it is the built-in management tenant. Set the tenant'sprimary_zoneto RANDOM.Test data size: 100 GB.
For step-by-step instructions, see OceanBase Database TPC-H testing.
Test results
The benchmark ran in SN mode with 100 GB of data and a concurrency of 16. Both hot run and cold run scenarios were tested:
Query |
100G |
100G |
|---|---|---|
| - | hot run(s) | cold run(s) |
| Concurrency | parallel(16) | parallel(16) |
| Q1 | 1.98s | 2.68s |
| Q2 | 0.20s | 1.06s |
| Q3 | 0.67s | 3.42s |
| Q4 | 0.58s | 3.67s |
| Q5 | 0.75s | 6.71s |
| Q6 | 0.03s | 0.65s |
| Q7 | 0.60s | 2.90s |
| Q8 | 0.85s | 8.10s |
| Q9 | 2.26s | 9.10s |
| Q10 | 0.65s | 3.60s |
| Q11 | 0.16s | 0.75s |
| Q12 | 0.27s | 1.26s |
| Q13 | 1.27s | 2.68s |
| Q14 | 0.19s | 0.46s |
| Q15 | 0.23s | 0.55s |
| Q16 | 0.48s | 0.55s |
| Q17 | 0.30s | 0.68s |
| Q18 | 2.07s | 2.89s |
| Q19 | 0.45s | 4.24s |
| Q20 | 0.23s | 1.56s |
| Q21 | 1.82s | 6.30s |
| Q22 | 0.28s | 0.73s |
| Total cost | 16.32s | 69.05s |
