Db2 11.1

For root installations, the database manager uses a formula to automatically adjust kernel parameter settings and eliminate the need for manual updates to these settings.

Before you begin

You must have root authority to modify kernel parameters.

Procedure

To update kernel parameters on Red Hat and SUSE Linux:

  1. Run the ipcs -l command to list the current kernel parameter settings.
  2. Analyze the command output to determine whether you have to change kernel settings or not by comparing the current values with the enforced minimum settings in the following table.
    IPC kernel parameter Enforced minimum setting
    kernel.shmmni (SHMMNI) 256 * <size of RAM in GB>
    kernel.shmmax (SHMMAX) <size of RAM in bytes>1
    kernel.shmall (SHMALL) 2 * <size of RAM in the default system page size>2
    kernel.sem (SEMMNI) 256 * <size of RAM in GB>
    kernel.sem (SEMMSL) 250
    kernel.sem (SEMMNS) 256 000
    kernel.sem (SEMOPM) 32
    kernel.msgmni (MSGMNI) 1 024 * <size of RAM in GB>
    kernel.msgmax (MSGMAX) 65 536
    kernel.msgmnb (MSGMNB) 65 536 3
    1. On 32-bit Linux operating systems, the enforced minimum setting for SHMMAX is limited to 4 294 967 295 bytes.
    2. SHMALL limits the total amount of virtual shared memory that can be allocated on a system. Each Db2® data server efficiently manages the amount of system memory it consumes, also know as committed memory. TheDb2 data server allocates more virtual memory than it commits to support memory preallocation and dynamic memory management. Memory preallocation benefits performance. Dynamic memory management is the process of growing and shrinking real memory usage within separate virtual shared memory areas. To support memory preallocation and dynamic memory management effectively, data servers frequently have to allocate more virtual shared memory on a system than the amount of physical RAM. The kernel requires this value as a number of pages.
    3. Load performance might benefit from a larger message queue size limit, which is specified in bytes by MSGMNB. You can view message queue usage by running the ipcs -q command. If the message queues are at capacity, or reaching capacity, during load operations, consider increasing the number of bytes the message queue size limit.

    The following text is an example of the ipcs command output with comments added after // to show what the parameter names are:

       # ipcs -l
    
       ------ Shared Memory Limits --------
    max number of segments = 4096 // SHMMNI
    max seg size (kbytes) = 32768 // SHMMAX
    max total shared memory (kbytes) = 8388608 // SHMALL
    min seg size (bytes) = 1 ------ Semaphore Limits --------
    max number of arrays = 1024 // SEMMNI
    max semaphores per array = 250 // SEMMSL
    max semaphores system wide = 256000 // SEMMNS
    max ops per semop call = 32 // SEMOPM
    semaphore max value = 32767 ------ Messages: Limits --------
    max queues system wide = 1024 // MSGMNI
    max size of message (bytes) = 65536 // MSGMAX
    default max size of queue (bytes) = 65536 // MSGMNB
    • Beginning with the first section on Shared Memory Limits, the SHMMAX limit is the maximum size of a shared memory segment on a Linux system. The SHMALL limit is the maximum allocation of shared memory pages on a system.

      • It is recommended to set the SHMMAX value to be equal to the amount of physical memory on your system. However, the minimum that is required on x86 systems is 268435456 (256 MB) and for 64-bit systems, it is 1073741824 (1 GB).
    • The next section covers the amount of semaphores available to the operating system. The kernel parameter semconsists of four tokens, SEMMSL, SEMMNS, SEMOPM, and SEMMNI. SEMMNS is the result of SEMMSL multiplied bySEMMNI. The database manager requires that the number of arrays (SEMMNI) be increased as necessary. Typically, SEMMNI must be twice the maximum number of agents expected on the system multiplied by the number of logical partitions on the database server computer plus the number of local application connections on the database server computer.
    • The third section covers messages on the system.
      • The MSGMNI parameter affects the number of agents that can be started. The MSGMAX parameter affects the size of the message that can be sent in a queue, and the MSGMNB parameter affects the size of the queue.
      • The MSGMAX parameter must be changed to 64 KB (that is, 65536 bytes), and the MSGMNB parameter must be increased to 65536.
  3. Modify the kernel parameters that you have to adjust by editing the /etc/sysctl.conf file. If this file does not exist, create it. The following lines are examples of what must be placed into the file:
    #Example for a computer with 16GB of RAM:
    kernel.shmmni=4096
    kernel.shmmax=17179869184
    kernel.shmall=8388608
    #kernel.sem=<SEMMSL> <SEMMNS> <SEMOPM> <SEMMNI>
    kernel.sem=250 1024000 32 4096
    kernel.msgmni=16384
    kernel.msgmax=65536
    kernel.msgmnb=65536

  4. Run sysctl with -p parameter to load in sysctl settings from the default file /etc/sysctl.conf:
       sysctl -p

  5. Optional: Have the changes persist after every reboot:
    • (SUSE Linux) Make boot.sysctl active.
    • (Red Hat) The rc.sysinit initialization script reads the /etc/sysctl.conf file automatically.

Kernel parameters for Db2 database server installation (Linux and UNIX)的更多相关文章

  1. FreeTDS-SQL Server在linux和unix下的免费驱动

    微软为MS SQL Server的连接和使用提供了很好的 驱动和 文档. 不幸的是,那只能在windows操作系统上使用. 所以对于Linux或者Unix,您需要寻找不同的方法来连接MS SQL Se ...

  2. Domino Server installation on Linux (Centos or Redhat) – something somewhere

    something somewhere welcome in there…:) Just another techki site howto / Linux / Lotus Domino 0 Domi ...

  3. 【转】Linux mysql停止失败的解决办法 Stopping MySQL database server mysqld [fail]

    root@ubuntu:/# /etc/init.d/mysql stop * Stopping MySQL database server mysqld [fail] root@ubuntu:/# ...

  4. Oracle EBS R12 (12.1.3) Installation Linux(64 bit)

    Oracle EBS R12 (12.1.3) Installation Linux(64 bit) Contents Objective. 3 1 Download & Unzip. 3 D ...

  5. SQL Server on Linux: How? Introduction: SQL Server Blog

    SQL Server Blog Official News from Microsoft’s Information Platform https://blogs.technet.microsoft. ...

  6. Acronis Server备份Linux系统

    Acronis Server备份Linux系统 前段时间用Acronis Disk Director Suite解决了Thinkpad笔记本在win7的分区问题(http://chenguang.bl ...

  7. 配置SQL Server on Linux(2)

    1. 前言 前一篇配置SQL Server on Linux(1),地址:http://www.cnblogs.com/fishparadise/p/8125203.html ,是关于更改数据库排序规 ...

  8. SQL Server On Linux:基于实际项目案例,总结功能支持情况及相关问题解决方案,讲如何快速完成迁移

    上个月,有个朋友问我说Sql Sever向Mysql迁移有什么好的经验分享,他们公司客户明确提出不再提供Windows服务器,现在计划Mysql迁移.我说Mysql迁移成本太高了,不妨可以了解一下SQ ...

  9. Howto: Connect MySQL server using C program API under Linux or UNIX

    From my mailbag: How do I write a C program to connect MySQL database server? MySQL database does su ...

随机推荐

  1. Qt下多线程日之类

    刚google到了,晚上回去试一下! 代码地址 https://gitorious.org/cutelogger/cutelogger/source/e3c2745c6c5f38896f87472e0 ...

  2. 【Android】详解Android 网络操作

    目录结构: contents structure [+] 判断网络 判断是否有网络连接 判断WIFI网络是否可用 判断MOBILE网络是否可用 获取当前网络连接的类型信息 监听网络 获取网络信息需要在 ...

  3. 团队工作效率分析工具gitstats

    如果你是团队领导,关心团队的开发效率和工作激情:如果你是开源软件开发者,维护者某个repo:又或者,你关心某个开源软件的开发进度,那么你可以试一试gitstats. gitstats 是一个git仓库 ...

  4. Effective Java 第三版——51. 仔细设计方法签名

    Tips 书中的源代码地址:https://github.com/jbloch/effective-java-3e-source-code 注意,书中的有些代码里方法是基于Java 9 API中的,所 ...

  5. MAC EI Capitan上更新系统自带SVN版本号(关闭SIP方能sudo rm)

    继昨晚之后.决定更新系统自带的svn.自带的svn版本号是1.7.看官网svn:http://www.wandisco.com/subversion/download#osx 最新版本号是1.9.13 ...

  6. 《网络编程》IPv4 与 IPv6 相互操作

    前言 因为互联网终端不断添加.IPv4 地址长度(32位)已不可以满足要求.所以出现了 IPv6地址(128位).可是现有应用程序大部分还是採用 IPv4 地址形式,所以必须解决 IPv4 与 IPv ...

  7. struts2 + urlrewrite 整合注意事项

    这几天业余时间在玩百度云,百度的云还是不错的,但是对于我这样的.NET程序员,有点不公平,没有.net虚机,不过也不是百度一家没有,基本都没有,有的都是那种开放云,自已在云端来部署安装软件的. 所以也 ...

  8. [转]Unity手游之路<十二>手游资源热更新策略探讨

    最近梳理了下游戏流程.恩,本来想写下,但是,还是看前辈的吧 版权声明: https://blog.csdn.net/janeky/article/details/17666409 上一次我们学习了如何 ...

  9. 【九天教您南方cass 9.1】 11 方格网土方计算

    同学们大家好,欢迎收看由老王测量上班记出品的cass9.1视频课程 我是本节课主讲老师九天. 我们讲课的教程附件也是共享的,请注意索取测量空间中. [点击索取cass教程]5元立得 (给客服说暗号:“ ...

  10. Bitmap用来做大数据处理

    MapReduce是一种编程模型,用于大规模数据集(大于1TB)的并行运算.概念"Map(映射)"和"Reduce(归约)" Bit-map空间压缩和快速排序去 ...