Shulou(Shulou.com)05/31 Report--

The main content of this article is "theoretical Analysis of ASM". Interested friends may wish to have a look. The method introduced in this paper is simple, fast and practical. Let's let the editor take you to learn the theoretical Analysis of ASM.

ASM Allocation Units

In the ASM disk group, the most basic space allocation unit is allocation unit, or AU for short. After initialization, each ASM disk is cut into an AU.

When a disk group is created, you can specify the size of the AU_SIZE by setting the attribute value of AU (after version 11.1). The size of the AU can be 1, 2, 4, 8, 16, 16, and 64 MB. If you do not specify the size of the AU, the default value is 1MB (4MB under Exadata).

AU size is a property of the disk group (not of the disk, not of the ASM instance), so each ASM disk group can have its own AU size value.

ASM Extents

One or more AU forms an extent, and one or more ASM extent forms an ASM file, so an ASM file is logically made up of extent.

We need to distinguish between physical extent and virtual extent, a virtual extent or extent set, which is composed of a physical extent in an external redundant disk group, at least 2 physical extent in an normal redundant disk group, and at least 3 physical extent in an high redundant disk group.

Before ASM 11.1, the size of extent was fixed. After ASM 11.1, variable extent appeared. The emergence of variable extent is to better support big data files, reduce the SGA requirements for ASM and database instances, and improve the performance of operations such as creating and opening files. The initialized extent size is equal to the AU_SIZE setting value of the disk group. As more and more extent are allocated for a file. The size of extent will increase by 4 or 16 times the AU_SIZE. This feature works automatically when creating a new file or resize, of course, the property values COMPATIBLE.ASM and COMPATIBLE.RDBMS of the ASM disk group should be set to greater than or equal to 11.1.

The extent size of a file changes as follows:

The size of the first 20000 extent set,extent of a file is equal to the set value of the AU_SIZE of the disk group.

The size of the next 20000 extent set,extent is equal to the AU_SIZE*4 of the disk group.

If the total number of extent set in a file is more than 40000, then the size of all subsequent extent is equal to the AU_SIZE*16 of the disk group.

This variable extent feature has an annoying BUG 8898852. For more information, please refer to MOS 965751.1.

ASM Mirroring

ASM's data mirroring function is used to protect the integrity of data, which is done by storing more copies of a piece of data on different disks. When an ASM disk group is created, the ASM administrator can specify how the disk group is mirrored:

External-does not provide mirror protection

Normal-2 copy

High-3 copy

The granularity of ASM mirroring is extent rather than disk or block,ASM mirroring is achieved by mirroring the extent of each ASM file that is made up. In ASM, we can specify the level of redundancy for each file. For example, for a file in a normal redundant disk group, each extent may be mirrored once (the default behavior), and another file, in the same disk group, may be mirrored twice, that is, three copies (assuming there are at least 3 failgroup in the disk group). In fact, ASM metadata files are made three copies in the normal redundant disk group, which also requires at least 3 ASMs in the disk group.

ASM Failgroups

An ASM disk group can be logically divided into a failgroup,failgroup that needs to be specified in the disk group creation. If we create a disk group without specifying failgroup, then ASM will automatically treat each disk as a failgroup, which may be different on Exadata. All disks from the same storage node under Exadata will be automatically put into a failgroup, even if you do not specify failgroup.

Normal redundant disk groups require at least 2 failgroup,high redundant disk groups require at least 3 failgroup,external redundant mode disk groups do not require failgroup.

When an extent is assigned to a file with two copies, ASM allocates a primary copy and a mirror copy,primary copy to be stored on one disk, while the mirror copy is stored on a different failgroup disk.

When adding disks to the ASM disk group, failgroup can specify them manually, and ASM will intelligently add disks to the correct failgroup.

At this point, I believe you have a deeper understanding of the "theoretical analysis of ASM". You might as well put it into practice. Here is the website, more related content can enter the relevant channels to inquire, follow us, continue to learn!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.