Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
2001-06-28
2004-11-16
Baderman, Scott (Department: 2184)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C711S118000
Reexamination Certificate
active
06820211
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
The present invention relates in general to data processing systems, and more particularly, data storage within data processing systems. Still more particularly, the present invention relates to a system and method for handling data requests within data processing systems.
2. Description of the Related Art
In modem computer systems, users frequently access common data stored in a redundant array of inexpensive disks (RAID) of a central networked computer. Redundancy is employed so that if a hard disk of the RAID fails, the data stored on the failed hard disk can be rebuilt utilizing data and parity data stored on the other hard disks of the RAID while the failed hard disk is taken off-line and repaired.
Hard disks within a RAID frequently are subject to numerous accesses ranging from relatively short data writes or reads to relatively lengthy updates of the configuration of the hard disk. During access to a disk, one or more users of the computer system may be unable to access the data on the disk. Such users typically wait until the hard disk completes the previous process until their data access requests can be fulfilled.
An example of a lengthy hard disk update that would restrict user access to data is a firmware update. Firmware updates may be required, for example, to resolve incompatibilities between components (e.g., hard disks) or to correct errors in the firmware that are not detected until after the release of the firmware.
An update to the firmware of a conventional hard disk cannot be performed when the drive is on-line. Because of this limitation, hard disk firmware updates are performed in a controlled operating system environment with the selected hard disk set to an off-line mode to prevent access to the hard disk. Utilizing a single threaded environment (e.g., disk operating system, or DOS) allows a hard disk firmware update application to assume control of the computer system and prevent other applications from accessing the hard disk during the firmware update.
Because many computer systems and networks employ multi-threaded operating systems (e.g., Windows NT), to perform a firmware update, an administrator typically shuts down the multi-threaded operating system, takes the computer system off-line, boots DOS, runs the hard disk update software, and waits for the update to complete. During the firmware update, the RAID is taken off-line, which prevents other users from accessing the data stored on the RAID.
Thus, it is often the case that users may request access to data stored on a hard disk that is unavailable due to an update, repair, or another access. The present invention recognizes that a system and method for servicing a data request to an unavailable hard disk without waiting for the update, access, or repair to be completed, would improve the overall of a data processing system.
SUMMARY OF THE INVENTION
To overcome the foregoing and additional limitations of the prior art, the present invention presents an improved system and method of servicing a data request to a storage media array. In accordance with a preferred embodiment of the present invention, the data processing system includes an adapter and a redundant array of inexpensive disks (RAID). The adapter includes an updating circuit and a request servicing circuit.
According to the method of the present invention, if a data read request to an unavailable particular storage medium is received from a system processor, a request servicing circuit of an adapter reconstructs the request data from data and parity data stored on other storage media in the array. However, if a data write request to an unavailable storage medium is received from the system processor, the actual data related to the data write request is stored in a temporary memory location, and the data write request is logged in a non-volatile memory. There is no built-in data redundancy in a first preferred embodiment of the present invention; thus, modified data is lost in the event of the failure of another storage medium during the update of the selected storage medium. However, a second preferred embodiment of the present invention incorporates a spare storage medium that preserves the modified data if another storage medium fails.
REFERENCES:
patent: 4135240 (1979-01-01), Ritchie
patent: 5522031 (1996-05-01), Ellis et al.
patent: 5600783 (1997-02-01), Kakuta et al.
patent: 5774643 (1998-06-01), Lubbers et al.
patent: 5875456 (1999-02-01), Stallmo et al.
patent: 6058489 (2000-05-01), Schultz et al.
patent: 6061750 (2000-05-01), Beardsley et al.
patent: 6067635 (2000-05-01), DeKoning et al.
patent: 6085333 (2000-07-01), DeKoning et al.
patent: 6098119 (2000-08-01), Surugucchi et al.
patent: 6243827 (2001-06-01), Renner, Jr.
patent: 6338126 (2002-01-01), Ohran et al.
patent: 6728833 (2004-04-01), Pruett et al.
patent: 11120113 (1999-04-01), None
Baderman Scott
Dillon & Yudell LLP
International Business Machines - Corporation
Lohn Joshua
LandOfFree
System and method for servicing requests to a storage array does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for servicing requests to a storage array, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for servicing requests to a storage array will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3358080