Why ASM?

. ASM is an Oracle provided Database/Cluster filesystem and Storage Volume manager, optimized for Oracle for any kind of workload (be it OLTP or DSS or Batch, sequential or random access), built-in with the database, with no additional cost.

. It follows SAME (Stripe and Mirror Everything) strategy, eliminating any database level IO hotspots, as the database files are striped equally across all the available disks in the allocated disk groups and are dynamically rebalanced with the addition and removal of disks online.

. ASM provides for better storage utilization and storage consolidation (ie allows multiple databases on the same Server or across multiple Servers to share storage through Oracle clusterware).

. It allows leveraging existing storage RAID technologies (Mirroring, RAID 5 etc) to provide for additional striping and redundancy.

. It can leverage IO multipathing technologies (like EMC Powerpath, Hitachi HDLM etc) to provide for better IO balancing and fault-tolerance.

. Makes use of Oracle ‘Kernel Service Direct File (ksfd)’ interface to perform KAIO (kernel asynchronous IO) against raw devices by-passing OS page cache, providing the best IO performance with no overhead from Operating system or conventional Logical Volume Managers.

关于‘ksfd’ 我们还可以看下面的解释:

Oracle performs 1MB I/O when possible

DBWR (the main server process) – Coalesces adjacent buffers to max size of 1MB

Kernel sequential file I/O (ksfq) – buffer size is set to 1MB by default

Archiver writes 1MB I/Os

Log Writer – Max I/O size is set to 1MB

Datapump,Dataguard,Rman and Transfer Packages also use ksfq

Kernel Service Direct File (ksfd) I/O is the central module providing disk I/O service to all its clients.the main interfaces provided allow client to create database file and perform I/O operations on them.Every read/write operation is done in multiples of the block size that is specified at file creation time

. Operating System Dependent (OSD) interface is used to access raw and file system files (used with ASM when ASMLIB is not configured) by default

. ASMLIB is disk access interface defined by Oracle and implemented by storage/system vendors. Oracle currently has an ASMLIB implementation for linux only . ASMLIB provides I/O enhancements for persistent binding, I/O optimization and ease of management

. Oracle Disk Manager (ODM) Interface defined by Oracle and implemented by storage/system vendors

ASM is one of the clients of ksfd. This module calls ksfd to issue I/O when ASM feature is enabled and there is NO ASMLIB library present. A database file is identified as an ASM managed file from the name pattern.

这里 我们还可以引申出对ASMLIB使用的考虑:

ASMLIB是一种基于Linux module,专门为Oracle Automatic Storage Management特性设计的内核支持库(kernel support library)。
理论上我们可以从ASMLIB API中得到的以下益处:
总是使用direct,async IO
ASMLIB API提供了传递如I/O优先级等元信息到存储设备的可能
虽然从理论上我们可以从ASMLIB中得到性能收益,但实践过程中这种优势是几乎可以忽略的,没有任何性能报告显示ASMLIB对比Linux上原生态的udev设备管理服务有任何性能上的优势。在Oracle官方论坛上有一篇讨论ASMLIB性能收益的帖子,你可以从中看到”asmlib wouldn’t necessarily give you much of an io performance benefit, it’s mainly for ease of management as it will find/discover the right devices for you, the io effect of asmlib is large the same as doing async io to raw devices.”的评论,实际上使用ASMLIB和直接使用裸设备(raw device)在性能上没有什么差别。
对于多路径设备(multipathing)需要在/etc/sysconfig/oracleasm-_dev_oracleasm配置文件中设置ORACLEASM_SCANORDER及ORACLEASM_SCANEXCLUDE,以便ASMLIB能找到正确的设备文件,具体可以参考Metalink Note
因为ASM INSTANCE使用ASMLIB提供的asm disk,所以增加了额外的层面
每次Linux Kernel更新,都需要替换新的ASMLIB包
使用ASMLIB创建的disk,其disk header并不会和普通的asm disk header有什么不同,仅仅是在头部多出了ASMLIB的属性空间。
我个人的观点是尽可能不要使用ASMLIB,当然这不是DBA个人所能决定的事情。另一方面这取决于个人习惯,在rhel 4的早期发行版本中没有提供udev这样的设备管理服务,这导致在rhel 4中大量的ASM+RAC组合的系统使用ASMLIB , 经网友指出udev 作为kernel 2.6的新特性被引入,在rhel4的初始版本中就已经加入了udev绑定服务,但是在rhel4时代实际udev的使用并不广泛(In Linux 2.6, a new feature was introduced to simplify device management and hot plug capabilities. This feature is called udev and is a standard package in RHEL4 or Oracle
Enterprise Linux 4 (OEL4) as well as Novell’s SLES9 and SLES10.)。如果是在RHEL/OEL 5中那么你已经有充分的理由利用udev而放弃ASMLIB。