Unlock instant, AI-driven research and patent intelligence for your innovation.

System and method for fault tolerant controller for network raid

a fault-tolerant controller and network raid technology, applied in error detection/correction, instruments, computing, etc., can solve problems such as data loss, inconsistent parity, and potential for inconsistent updates of physically distributed redundant data, so as to reduce the space required for tags

Inactive Publication Date: 2007-12-20
INT BUSINESS MASCH CORP
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

Ensures data consistency and fault tolerance by instantiating a new coordinator to repair inconsistencies, reducing the risk of data loss and maintaining data integrity across RAID systems.

Problems solved by technology

As recognized herein, however, if these stripes are not repaired before a subsequent failure(s) then data loss will occur.
As understood herein, a challenge to allowing clients to share access to data is the potential for inconsistent updates to the physically distributed redundant data.
But if two clients happen to concurrently write, then a wrong interleaving of the two phases of each write can lead to an inconsistent parity, which can occur even when the clients are writing to disjoint logical address ranges.
Another source of inconsistent parity occurs when a client fails in the middle of its second phase of a write operation.
In both cases, the end result is that the data and parity blocks are inconsistent.
If a storage device in the system fails before this inconsistency is repaired, then data loss will occur.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • System and method for fault tolerant controller for network raid
  • System and method for fault tolerant controller for network raid
  • System and method for fault tolerant controller for network raid

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0021] Referring initially to FIG. 1, a data storage system is shown, generally designated 10, in which large storage volumes such as storage area networks, which may be established by multiple disk drives in a system such as but not limited to a RAID-5 system, are divided into smaller units 12 referred to herein as “extents”. As shown, each extent 12 is associated with a respective coordinator 14. All data from plural client host device 16 that is to be stored in the extent managed by a coordinator 14 passes through the coordinator 14. A coordinator 14 may be instantiated by any one of the plural drive controllers associated with respective storage devices 18 (e.g., disk drives) in the extent.

[0022] As intended herein, a coordinator 14 serializes all accesses to its associated extent 12, and specifically serializes potentially interfering file updates to physical storage that spans multiple storage devices 18. Also, the coordinator 14 for an extent 12 implements the redundancy, ca...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A fault-tolerant and efficient way of deducing a set of inconsistent stripes for a network RAID protocol, wherein clients forward input / output (I / O) to a particular controller device called the coordinator, which executes RAID logic and which sends out device IOs to the relevant storage devices. If the coordinator fails then a new coordinator reconstructs its state from the storage devices.

Description

FIELD OF THE INVENTION [0001] The present invention relates generally to redundant data storage systems such as redundant arrays of independent disks (RAID) systems. BACKGROUND [0002] RAID systems that span multiple controllers are referred to as “network RAID”. The purpose of RAID systems is to provide data redundancy, so that a single RAID module failure will not result in lost data. When a RAID module fails abruptly (either due to software error or hardware failure) it can leave inconsistent parity / mirror stripes, which can be repaired. As recognized herein, however, if these stripes are not repaired before a subsequent failure(s) then data loss will occur. [0003] In many RAID systems, client devices, also referred to herein as “hosts”, are connected to multiple storage controllers. Data redundancy is maintained across controllers in accordance with RAID principles, with software being executed on each controller to coordinate layout and access to data and its recovery on compone...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(United States)
IPC IPC(8): G06F12/16
CPCG06F11/2089
Inventor FLEINER, CLAUDIO MATTHIASGOLDING, RICHARD ANDREWKENCHAMMANA-HOSEKOTE, DEEPAK R.ZAKL, OMER AHMED
Owner INT BUSINESS MASCH CORP