Browse Prior Art Database

Method and System for Augmenting RAID 5 protection with RAID 6 Protection

IP.com Disclosure Number: IPCOM000243137D
Publication Date: 2015-Sep-17
Document File: 3 page(s) / 68K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method and system for opportunistically using idle, unused storage subsystem resources to augment Redundant Array of Independent Disks (RAID) 5 protection with RAID 6 protection.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 3

Method and System for Augmenting RAID 5 protection with RAID 6 Protection


Storage subsystems often support RAID (Redundant Array of Independent Disks). The purpose of RAID is to increase availability or performance of storage through a combination of distributing and duplicating data over a collection of storage devices, which may be disks or flash modules.

RAID 5 is a particular form of RAID where a logically contiguous fixed-size set of blocks (the 'data blocks') are distributed evenly over a set of devices, and augmented by a parity block which is stored on a separate device. The logically contiguous set of blocks is usually known as a 'stripe'. Contents of a parity block are calculated from the contents of the data blocks in such a way that original data can be reconstructed in case of corruption of any single data block. Each of the data and parity blocks are stored on different devices, so that the original data can be reconstructed in case of a single device failure. RAID 6 augments RAID 5 with an additional parity block per stripe, allowing the original data to be reconstructed in case of a double device failure.

Storage subsystems often have idle, unused resources. An example is an active spare, also referred to as a "hot spare". Conventionally, active spares are only used when a drive or flash module fails. Data from the failed module is reconstructed on the active spare, to maintain the same level of data protection that existed before the failure. Another

example of an idle resource would be the CPU capacity on a lightly-used storage subsystem.

A method and system is disclosed for opportunistically using idle, unused storage subsystem resources to augment Redundant Array of Independent Disks (RAID) 5 protection with RAID 6 protection.

In accordance with the system disclosed here, RAID 5 is handled by the storage subsystem as before, except that an additional flag is maintained for each stripe that indicates if the RAID 6 parity block is valid. When a block is written in a stripe, the RAID 5 parity block is updated as usual, and the RAID 6 flag is set to 'invalid'.

As available, idle CPU capacity is used in the background to calculate and write RAID 6 parity blocks. When a RAID 6 parity block is written, the RAID 6 parity block flag for that stripe is changed from 'invalid' to 'valid'. The RAID 6 parity blocks are maintained on an active spare.

When a single device failure occurs, data can be reconstructed using RAID 5 parity as before. The storage subsystem chooses an active spare on which to reconstruct the failed device. This...