Oracle Automatic Storage Management (ASM) is a key component in Oracle Exadata systems, managing how grid disks are grouped, accessed, and optimized for performance and fault tolerance.

This article explains the basics of ASM disk groups, how to configure them, and best practices for reliability and speed.

What Are ASM Disk Groups?

• ASM Disk Group: A logical collection of grid disks managed by Oracle ASM.
• Grid Disk Naming Pattern:

o/cell_IPaddress/griddisk_name

• cell_IPaddress → IP of the Exadata storage server.
• griddisk_name → Name of the grid disk.
• Prefix o/ is always used.

Naming Recommendation: Use similar names for disk groups and their grid disks for clarity.

Key Considerations When Creating Disk Groups

• Default Names:
  • Disk name = grid disk name (recommended).
  • Failure group name = cell name (recommended).
• Wildcards: Use * to include multiple disks easily. Example:

CREATE DISKGROUP reco HIGH REDUNDANCY DISK 'o/*/DATA*';

Failure Groups:

• If not specified, ASM assigns disks to failure groups based on their cell.
• If a cell is renamed, specify the original failure group name to keep consistency.

Smart Scan Requirement:

• All disks in a group must be Exadata grid disks (no mixing with conventional disks).

Performance Optimization

• Fast Disk Scans:
  • Ensure at least 4 MB contiguous space for segments.
  • Set ASM allocation unit size (AU_SIZE) to 4 MB.
• Example Command:

CREATE DISKGROUP data NORMAL REDUNDANCY 
DISK 'o/*/data_CD*'
ATTRIBUTE 'compatible.rdbms' = '11.2.0.2', 
          'compatible.asm' = '11.2.0.3',
          'content.type' = 'data',
          'cell.smart_scan_capable' = 'TRUE',
          'au_size' = '4M';

Enhancing Fault Tolerance with Content Types

• Content Type Attribute (content.type): Defines the role of a disk group.
• Options:
  • data → For DATA and SPARSE disk groups.
  • recovery → For RECO disk group.
  • system → For DBFS_DG disk group.
• Benefit: Distributes data differently across disks, reducing risk of data loss in double or triple failures.
• Example:

ALTER DISKGROUP reco SET ATTRIBUTE 'content.type'='recovery';
ALTER DISKGROUP reco REBALANCE POWER preferred_power_setting;

Checking Content Type

Run this query to verify disk group attributes:

SELECT dg.name, a.value 
FROM v$asm_diskgroup dg, v$asm_attribute a 
WHERE dg.group_number=a.group_number 
AND a.name='content.type' 
AND (dg.name LIKE 'DATA%' OR dg.name LIKE 'RECO%' OR dg.name LIKE 'DBFS_DG%');

Sample Output:

Best Practices

• Use default disk and failure group names for simplicity.
• Always set AU_SIZE = 4 MB for faster scans.
• Assign proper content.type values to enhance fault tolerance.
• Ensure all disks in a group are Exadata grid disks.
• Rebalance after changing attributes to maintain redundancy.

In Summary

Oracle ASM disk groups in Exadata provide structured storage management, high performance, and strong fault tolerance. By using proper naming, wildcards, allocation units, and content type attributes, administrators can ensure efficient data access and resilience against failures.