Method and apparatus for hierarchical management of data storage elements in an array storage device

• US 5,566,316 A
• Filed: 02/23/1996
• Issued: 10/15/1996
• Est. Priority Date: 02/10/1994
• Status: Expired due to Term

• Current Assignee: Storage Technology Corp (Oracle Corporation)
• Original Assignee: Storage Technology Corp (Oracle Corporation)
• Inventors: Nguyen, Thai, Fechner, Jimmy R., Raymond, Robert M.
• Primary Examiner(s): SHAH, ALPESH
• Application Number:
  US08/606,223
• Time in Patent Office: 235 Days
• Field of Search: 395/182.03-182.06, 395/404, 395/439, 395/441, 395/444, 395/497.01, 395/800, 395/827, 395/835, 395/856, 371/2.2, 371/10.1, 371/11.4, 371/40.1, 365/189.04, 365/230.01, 365/230.03
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1662 the resynchronized component or unit being a persistent storage device re-synchronization of failed mirror storage G06F11/2082; rebuild or reconstruction of parity RAID storage G06F11/1008
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G11C 29/74 using duplex memories, i.e. using dual copies

Abstract

The data storage subsystem is implemented using redundancy groups of data storage modules, at least one module of which comprises a plurality of data storage elements. In this manner, the one dimensional data storage module redundancy group is expanded in another dimension by the use of a plurality of storage elements to implement each of the data storage modules. Enhanced performance is obtained by having more actuators concurrently accessing data. The reconstruction of data due to an equipment failure is therefore localized to a data storage element rather than requiring the replacement of an entire data storage module.

Claims

1. A data storage subsystem comprising:

a plurality of data storage elements;

a plurality of cluster controllers each connected to a subset of said plurality of data storage elements for storing information;

means within at least one of said plurality of cluster controllers for controlling a first level redundancy group, wherein said first level redundancy group consists of the connected subset of data storage elements, and wherein N of said data storage elements of said connected subset are used by the connected cluster controller to store data, and wherein M of said data storage elements of said connected subset are used by the connected cluster controller to store redundancy information, and wherein at least one of said connected subset of data storage elements is used by the connected cluster controller as a spare data storage element;

first level correction means within each of said at least one of said plurality of cluster controllers for correcting errors detected in said first level redundancy group by using data and redundancy information stored in said data storage elements of said connected subset;

an array controller connected to said plurality of cluster controllers for controlling a second level redundancy group, wherein said second level redundancy group consists of said plurality of cluster controllers, wherein N of said plurality of cluster controllers are used by said array controller to store data, wherein M of plurality of cluster controllers are used by array controller to store redundancy information, and wherein at least one of said plurality of cluster controllers is used by array controller as a spare cluster controller; and

second level correction means within said array controller for correcting errors detected in said second level redundancy group by using data and redundancy information stored in said plurality of cluster controllers.

2. The data storage subsystem of claim 1 wherein said first level correction means includes:

means for determining that an error detected in said first level redundancy group corresponds to a failed data storage element;

rebuild correction means, responsive to a determination that said errors do not correspond to a failed data storage element, for correcting errors detected in said first level redundancy group by rebuilding data corresponding to said detected errors; and

replacement correction means, responsive to a determination that said errors correspond to a failed data storage element, for correcting errors detected in said first level redundancy group by replacement of said failed data storage element with a spare data storage element from said first level redundancy group and rebuilding of data on said spare data storage element.

3. The data storage subsystem of claim 1 wherein said second level correction means includes:

means for determining that an error detected in said second level redundancy group corresponds to a failed cluster controller;

rebuild correction means, responsive to a determination that said errors do not correspond to a failed cluster, for correcting errors detected in said second level redundancy group by rebuilding data corresponding to said detected errors; and

replacement correction means, responsive to a determination that said errors correspond to a failed cluster controller, for correcting errors detected in said second level redundancy group by replacement of said failed cluster controller with a spare cluster controller from said second level redundancy group and rebuilding of data on said spare cluster controller.