matrixorigin / matrixorigin/matrixone
[Bug]: sharding performance is poor.
- Dominant language
- Go
- Stars
- 1.9k
- Forks
- 311
- Avg merge
- 1d 3h
- Merged PRs (30d)
- 768
Description
### Is there an existing issue for the same bug?
- [X] I have checked the existing issues.
### Branch Name
main
### Commit ID
b0855d26f
### Other Environment Information
```Markdown
- Hardware parameters:
- OS type:
- Others:
```
### Actual Behavior
**5cn,每个cn 100并发sharding的测试结果:**
**1cn,100并发无sharding的测试结果:**
### Expected Behavior
_No response_
### Steps to Reproduce
```Markdown
带sharding的测试结果,启动集群配置中,添加cn和dn的配置,cn的副本数配置为5:
cn配置:
[cn.shardservice]
enable = true
dn配置:
[dn.shardservice]
enable = true
max-cn-freeze-time = "5s"
max-schedule-tables = 4
git clone https://github.com/matrixorigin/matrixone.git
cd mo-sysbench
cleanup:
sysbench --mysql-host=172.16.22.5 --mysql-port=6001 --mysql-user=dump --mysql-password=111 oltp_update_index.lua --mysql-db=sysbench_db --tables=1 --table_size=100000 --threads=100 --time=30 --report-interval=10 cleanup
prepare(prepare前需要先创建db:sysbench_db):
sysbench --mysql-host=172.16.22.5 --mysql-port=6001 --mysql-user=dump --mysql-password=111 oltp_update_index.lua --mysql-db=sysbench_db --tables=1 --table_size=100000 --threads=100 --time=30 --report-interval=10 prepare
git clone https://github.com/matrixorigin/mo-load.git
根据5个cn,查找cn对应的uuid,以及table_id,将每个cn的范围设置成对应的shard的范围,然后使用5个mo-load工具同时测试tps,将5个tps相加即为总tps,如测试脚本如有需要,联系我
不带sharding的测试结果
启动集群,cn的副本数配置为1:
git clone https://github.com/matrixorigin/mo-load.git
cd mo-load
vi mo.yaml(修改ip)
cd cases/sysbench/sysbench/update_pk_10_100000_prepare
vi replace.yml(修改shard范围)
./start.sh -c cases/sysbench/update_pk_10_100000_prepare -t 100 -d 1 -b sysbench_db
```
### Additional information
_No response_
Contributor guide
Assessment
This issue has not been assessed yet.