You are viewing a plain text version of this content. The canonical link for it is here.
Posted to notifications@shardingsphere.apache.org by GitBox <gi...@apache.org> on 2020/04/27 09:39:51 UTC

[GitHub] [shardingsphere] ChuckPerry opened a new issue #5345: [performance test] proxy is too slow with sysbench

ChuckPerry opened a new issue #5345:
URL: https://github.com/apache/shardingsphere/issues/5345


   ## Question
   I run sharding-proxy with no sharding mode. there's only 100+ qps , and with same parameters, there's 8000+ qps when I connect mysql directly.
   ![Snipaste_2020-04-27_17-28-10](https://user-images.githubusercontent.com/10229398/80356885-b3c4e800-88ac-11ea-9fec-fbe22d007124.jpg)
   ![Snipaste_2020-04-27_17-29-42](https://user-images.githubusercontent.com/10229398/80356903-b9bac900-88ac-11ea-8d22-ac4dcf66cbb3.jpg)
    
   Environment:
   - mysql 5.7.29
   - sharding-proxy 4.0.1
   - centos 7.4.18
   - jdk 1.8+
   
   Config files:
   - server.yaml
   `#orchestration:`
   `#  name: orchestration_ds`
   `#  overwrite: true`
   `#  registry:`
   `#    type: zookeeper`
   `#    serverLists: localhost:2181`
   `#    namespace: orchestration`
   `#`
   `authentication:`
   `  users:`
   `    root:`
   `      password: root`
   `      authorizedSchemas: test_db`
   `    sharding:`
   `      password: sharding`
   `      authorizedSchemas: sharding_db`
   `#`
   `props:`
   `  max.connections.size.per.query: 1`
   `  acceptor.size: 16  # The default value is available processors count * 2.`
   `  executor.size: 16  # Infinite by default.`
   `  proxy.frontend.flush.threshold: 512  # The default value is 128.`
   `#    # LOCAL: Proxy will run with LOCAL transaction.`
   `#    # XA: Proxy will run with XA transaction.`
   `#    # BASE: Proxy will run with B.A.S.E transaction.`
   `  proxy.transaction.type: XA`
   `  proxy.opentracing.enabled: false`
   `  query.with.cipher.column: true`
   `  sql.show: false`
   - config-master_slave.yaml
   `schemaName: test_db`
   `dataSources:`
   `  master_ds:`
   `    url: jdbc:mysql://10.10.10.103:3306/test_db?serverTimezone=UTC&useSSL=false`
   `    username: user`
   `    password: xxxxxx`
   `    connectionTimeoutMilliseconds: 30000`
   `    idleTimeoutMilliseconds: 60000`
   `    maxLifetimeMilliseconds: 1800000`
   `    maxPoolSize: 100`
   `  slave_ds_0:`
   `    url: jdbc:mysql://10.10.10.102:3306/test_db?serverTimezone=UTC&useSSL=false`
   `    username: user`
   `    password: xxxxxx`
   `    connectionTimeoutMilliseconds: 30000`
   `    idleTimeoutMilliseconds: 60000`
   `    maxLifetimeMilliseconds: 1800000`
   `    maxPoolSize: 100`
   `  slave_ds_1:`
   `    url: jdbc:mysql://10.10.10.101:3306/test_db?serverTimezone=UTC&useSSL=false`
   `    username: user`
   `    password: xxxxxx`
   `    connectionTimeoutMilliseconds: 30000`
   `    idleTimeoutMilliseconds: 60000`
   `    maxLifetimeMilliseconds: 1800000`
   `    maxPoolSize: 100`
   `#masterSlaveRule:`
   `#  name: ms_ds`
   `#  masterDataSourceName: master_ds`
   `#  slaveDataSourceNames:`
   `#    - slave_ds_0`
   `#    - slave_ds_1`
   Is there any place that I'm wrong?


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [shardingsphere] tuohai666 commented on issue #5345: [performance test] proxy is too slow with sysbench

Posted by GitBox <gi...@apache.org>.
tuohai666 commented on issue #5345:
URL: https://github.com/apache/shardingsphere/issues/5345#issuecomment-619902784


   Don't worry, you didn't do anything wrong.
   The performance issue in 4.0.1 will be resolved in 4.1.0 which is in releasing process(under voting now).
   The performance will be improved by two aspect:
   1. New parser engine that will improve parse performance for long SQL by more than 10000%.
   2. Fixed several C API related issues: #4358, #4355, #4255, #3968, #4501
   
   You can compile the codes of 4.1.0 in https://github.com/apache/shardingsphere/releases/tag/4.1.0.
   Or just wait the new release for about one week.


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



[GitHub] [shardingsphere] rouxiaoxiao commented on issue #5345: [performance test] proxy is too slow with sysbench

Posted by GitBox <gi...@apache.org>.
rouxiaoxiao commented on issue #5345:
URL: https://github.com/apache/shardingsphere/issues/5345#issuecomment-738521217


   > Don't worry, you didn't do anything wrong.
   > The performance issue in 4.0.1 will be resolved in 4.1.0 which is in releasing process(under voting now).
   > The performance will be improved by two aspect:
   > 
   > 1. New parser engine that will improve parse performance for long SQL by more than 10000%.
   > 2. Fixed several C API related issues: #4358, #4355, #4255, #3968, #4501
   > 
   > You can compile the codes of 4.1.0 in https://github.com/apache/shardingsphere/releases/tag/4.1.0.
   > Or just wait the new release for about one week.
   
   I have the same problem and the speed is not improved after upgrading to version 4.1.0.The following are my details:
   sql cost:110ms
   jdbc cost: 226ms
   sharding jdbc 4.0.1: 2883ms
   sharding jdbc 4.1.0:2930ms


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org