Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reexamination Certificate
1998-12-02
2002-08-06
Myers, Paul R. (Department: 2181)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C714S770000
Reexamination Certificate
active
06430701
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to data recording and reproducing method and apparatus using a plurality of data recording and reproducing units, and a computer-readable recording medium. More specifically, the present invention is directed to data recording and reproducing method and apparatus using a plurality of recording and reproducing means that, at the time of data recording, produces pieces of second data by dividing supplied first data and adds an error correction code to the pieces of second data, allocates and records the pieces of second data and the error correction code to and in the plurality of recording and reproducing units, respectively, and that, at the time of data reproduction, reproduces the first data by synthesizing the pieces of second data read from the plurality of recording and reproducing units.
Known as a highly reliable data recording and reproducing apparatus having a large storage capacity is, e.g., a disk array (RAID or Redundant Array of Inexpensive Disks) unit that controls a plurality of data recording and reproducing units such as hard disk drives (hereinafter referred to as “HDD(s)”) arranged in parallel to one another as if the HDDs were handled as a single disk unit.
To record data using such a disk array unit, the data to be recorded is divided, and the divided pieces of data are sequentially and cyclically allocated to and recorded in the plurality of HDDS. Further, an operation is performed to obtain a parity based on the divided pieces of data recorded in the plurality of HDDs. In a RAID level 3 or 4 system, the obtained parity is recorded in a single fixed HDD. In a RAID level 5 system, the obtained parities are regularly distributed to the respective HDDs.
To reproduce the data from the divided pieces of data recorded in the plurality of HDDS, the pieces of data read from the plurality of HDDs are subjected to an error detection correction using the recorded parity, and the resultant pieces of data are synthesized and outputted. Further, when a fault occurs at the disk array unit, a rebuilt process is performed. In the rebuild process, a faulty HDD is replaced with a new HDD and a piece of data recorded in the faulty HDD is written to the new HDD using the parity.
Since the disk array unit records data using a plurality of HDDs as described above, a large volume of data can be recorded. Further, since parity is used to make error detection correction and the rebuild process is performed to write back a piece of data to a newly replaced HDD, high reliability can be assured (prior art (1)). One such example is disclosed in U.S. Pat. No. 4,870,643.
Further, such a disk array unit reserves an auxiliary HDD. Therefore, when a fault occurs at an HDD, a so-called “hot standby” technique using the auxiliary HDD is utilized (prior art (2)). That is, a piece of data recorded in the faulty HDD is recovered by using pieces of data and a parity recorded in the faultless HDDS, and the recovered piece of data is recorded in the auxiliary HDD. In this technique, the data is thereafter recorded or reproduced by using the pieces of data recorded in the faultless HDDS and the piece of data recorded in the auxiliary HDD. Further, when the faulty HDD is replaced with a new HDD, the replaced HDD is reserved as the auxiliary HDD.
Since such a disk array unit records data in a plurality of HDDs as described above, a large volume of data can be recorded. In addition, since a parity is used to make error detection correction and an auxiliary HDD is reserved, high reliability can be assured.
SUMMARY OF THE INVENTION
By the way, the prior art (1) addresses the following problem. When a first fault occurs at the disk array unit, and if a second fault then occurs at another HDD during a period between the replacement of the faulty HDD with a new HDD and the completion of a rebuild process, then a piece of data in the second faulty HDD cannot be recovered by using a parity. Therefore, the recorded data cannot be read correctly.
Further, the prior art (2) in which the disk array unit has an auxiliary HDD also addresses the following problem. When a second fault occurs at an HDD during a period in which data is being recorded or reproduced using the auxiliary HDD and in which a first faulty HDD has not yet been replaced with a new HDD, a piece of data in the second faulty HDD can be recovered by using a parity. However, if a third fault occurs at an HDD, a piece of data in the third faulty HDD cannot be recovered by using a parity. Therefore, the recorded data cannot be read correctly.
An object of the present invention is to provide data recording and reproducing method and apparatus that can overcome the aforementioned problems addressed by the conventional data recording and reproducing method and apparatus.
Another object of the present invention is to provide data recording and reproducing method and apparatus that can improve reliability.
A first aspect of the present invention provides a data recording and reproducing method using a plurality of recording and reproducing means in which, at the time of data recording, pieces of second data are produced by dividing supplied first data and an error correction code is generated for the pieces of second data, and the pieces of second data and the error correction code are allocated to and recorded in the plurality of recording and reproducing means, respectively, and in which, at the time of data reproduction, the first data is reproduced by synthesizing the pieces of second data read from the plurality of recording and reproducing means. The method comprises the steps of: when a fault occurs at one of the plurality of recording and reproducing means (hereinafter referred to as the “faulty recording and reproducing means”), if one of the pieces of second data is recorded in the faulty recording and reproducing means, reproducing the first data by reading the pieces of second data and the error correction code recorded in the plurality of recording and reproducing means excluding the faulty recording and reproducing means (hereinafter referred to as the “faultless recording and reproducing means”), and if one of the pieces of second data is not recorded in the faulty recording and reproducing means, reproducing the first data by reading the pieces of second data recorded in the plurality of faultless recording and reproducing means; and producing new pieces of second data by dividing the reproduced first data and generating a new error correction code, and allocating and recording the new pieces of second data and the new error correction code to and in the plurality of faultless recording and reproducing means, respectively.
The first aspect of the present invention further provides a data recording and reproducing apparatus using a plurality of recording and reproducing means, which apparatus comprises: a plurality of recording and reproducing means; data processing means that, at the time of data recording, produces pieces of second data by dividing supplied first data and generates an error correction code, and allocates and records the pieces of second data and the error correction code to and in the plurality of recording and reproducing means, respectively, and that, at the time of data reproduction, reproduces the first data by synthesizing the pieces of second data read from the plurality of recording and reproducing means; and operation control means that, when a fault occurs at one of the plurality of recording and reproducing means (hereinafter referred to as the “faulty recording and reproducing means”), controls the data processing means, if one of the pieces of second data is recorded in the faulty recording and reproducing means, to reproduce the first data from the pieces of second data and the error correction code recorded in the plurality of recording and reproducing means excluding the faulty recording and reproducing means (hereinafter referred to as the “faultless recording and reproducing means”), and that controls the data processing means, if one of the piece
Abel David B.
Aiwa Co. Ltd.
Myers Paul R.
Squire Sanders & Dempsey L.L.P.
LandOfFree
Data recording and reproducing method and apparatus using... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Data recording and reproducing method and apparatus using..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Data recording and reproducing method and apparatus using... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2956545