• 并行进程


    SQL> show parameter parallel_min

     

    NAME                     TYPE    VALUE

    ------------------------------------ ----------- ------------------------------

    parallel_min_percent             integer 0

     

    SQL> show parameter parallel_max

     

    NAME                     TYPE    VALUE

    ------------------------------------ ----------- ------------------------------

    parallel_max_servers             integer 960

     

    SQL> select * from v$px_process_sysstat;

     

    STATISTIC                             VALUE

    ------------------------------------------------------------ ----------

    Servers In Use                                0

    Servers Available                            64

    Servers Started                             308

    Servers Shutdown                            244

    Servers Highwater                            64

    Servers Cleaned Up                            0

    Server Sessions                          226823

    Memory Chunks Allocated                         105

    Memory Chunks Freed                          87

    Memory Chunks Current                            18

    Memory Chunks HWM                            23

     

    STATISTIC                             VALUE

    ------------------------------------------------------------ ----------

    Buffers Allocated                       1771708

    Buffers Freed                           1771708

    Buffers Current                               0

    Buffers HWM                            1566

     

    15 rows selected.

     

    Servers In Use: 正在使用的并行进程的数量

     

    Servers Available:可用的并行进程的数量

     

    执行SQL:

    SQL> select /*+ parallel(a 8)*/ count(*) from T_PM_ACCT_DTL a;

    对应8个进程

    祖宗进程为384

    那么加上group by order by呢?

    SQL> select /*+ parallel(a 8)*/ count(*) from T_PM_ACCT_DTL a group by a.data_date order by a.data_date;

    可以看到已经使用了16个

  • 相关阅读:
    test deploy
    SpringBoot+Spring Session+Redis实现Session共享及踩坑记录
    登录接口取cookie
    pycharm设置自动调节字体大小
    ClickHouse高可用集群
    clickhouse 常用命令(三)
    clickHouse常用命令(一)
    ClickHouse表引擎
    clickhouse基本数据类型
    手写滚动条设计----直接粘贴
  • 原文地址:https://www.cnblogs.com/zhaoyangjian724/p/3797956.html
Copyright © 2020-2023  润新知