Error detection/correction and fault detection/recovery – Data processing system error or fault handling – Reliability and availability
Reissue Patent
1995-11-15
2002-03-19
Decady, Albert (Department: 2133)
Error detection/correction and fault detection/recovery
Data processing system error or fault handling
Reliability and availability
C711S100000
Reissue Patent
active
RE037601
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field:
This invention relates in general to methods and systems for maintaining continued availability of datasets in external storage associated with accessing data processing systems, and in particular the present invention relates to backup copying of records in external storage concurrent with a dramatically shortened suspension of data processing system application execution occasioned by such copying. Still more particularly, the present invention relates to a method and system for incremental backup copying of records in a data processing system which minimizes still further the suspension of data processing system application execution during such copying.
2. Description of the Related Art:
A modern data processing system must be prepared to recover, not only from corruptions of stored data which occur as a result of noise bursts, software bugs, media defects, and write path errors, but also from global events, such as data processing system power failure. The most common technique of ensuring the continued availability of data within a data processing system is to create one or more copies of selected datasets within a data processing system and store those copies in a nonvolatile environment. This so-called “backup” process occurs within state-of-the-art external storage systems in modern data processing systems.
Backup policies are implemented as a matter of scheduling. Backup policies have a space and time dimension which is exemplified by a range of datasets and by the frequency of backup occurrence. A FULL backup requires the backup of an entire range of a dataset, whether individual portions of that dataset have been updated or not. An INCREMENTAL backup copies only that portion of the dataset which has been updated since a previous backup, either full or incremental. The backup copy thus created represents a consistent view of the data within the dataset as of the time the copy was created.
Of course, those skilled in the art will appreciate that as a result of the process described above, the higher the backup frequency, the more accurately the backup copy will mirror the current state of data within a dataset. In view of the large volumes of data maintained within a typical state-of-the-art data processing system backing up that data is not a trivial operation. Thus, the opportunity cost of backing up data within a dataset may be quite high on a large multiprocessing, multiprogramming facility, relative to other types of processing.
Applications executed within a data processing system are typically executed in either a batch (streamed) or interactive (transactional) mode. In a batch mode, usually one application at a time executes within interruption. Interactive mode is characterized by interrupt driven multiplicity of applications or transactions.
When a data processing system is in the process of backing up data in either a streamed or batch mode system, each process, task or application within the data processing system is affected. That is, the processes supporting streamed or batch mode operations are suspended for the duration of the copying. Those skilled in the art will recognize that this event is typically referred to as a “backup window.” In contrast to batch mode operations, log based or transaction management applications are processed in the interactive mode. Such transaction management applications eliminate the “backup window” by concurrently updating an on-line dataset and logging the change. However, this type of backup copying results in a consistency described as “fuzzy.” That is, the backup copy is not a precise “snapshot” of the state of a dataset/data base at a single point in time. Rather, a log comprises an event file requiring further processing against the database.
A co-pending U.S. patent application Ser. No. 07/385,647, filed Jul. 25, 1989, entitled A Computer Based Method For Dataset Copying Using an Incremental Backup Policy, illustrates backup in a batch mode system utilizing a modified incremental policy. A modified incremental policy copies only new data or data updates since the last backup. It should be noted that execution of applications within the data processing system are suspended during copying in this system.
As described above, to establish a prior point of consistency in a log based system, it is necessary to “repeat history” by replaying the log from the last check point over the datasets or database of interest. The distinction between batch mode and log based backup is that the backup copy is consistent and speaks as of the time of its last recordation, whereas the log and database mode require further processing in the event of a fault, in order to exhibit a point in time consistency.
U.S. Pat. No. 4,507,751, Gawlick et al., entitled Method and Apparatus for Logging Journal Data Using a Write Ahead Dataset, issued Mar. 25, 1985, exemplifies a transaction management system wherein all transactions are recorded on a log on a write-ahead dataset basis. As described within this patent, a unit of work is first recorded on the backup medium (log) and then written to its external storage address.
Co-pending U.S. patent application Ser. No. 07/524,206, filed May 16, 1990, entitled Method and Apparatus for Executing Critical Disk Access Commands, teaches the performance of media maintenance on selected portions of a tracked cyclic operable magnetic media concurrent with active access to other portions of the storage media. The method described therein requires the phased movement of customer data between a target track to an alternate track, diversion of all concurrent access requests to the alternate track or tracks and the completion of maintenance and copy back from the alternate to the target track.
Requests and interrupts which occur prior to executing track-to-track customer data movement result in the restarting of the process. Otherwise, requests and interrupts occurring during execution of the data movement view a DEVICE BUSY state. This typically causes a requeueing of the request.
It should therefore be apparent that a need exists for a method and system whereby the maximum availability of application execution within a data processing system is maintained while creating backup copies which exhibit a consistent view of data within an associated database, as of a specific time.
SUMMARY OF THE INVENTION
It is therefore one object of the present invention to provide an improved method and system for maintaining continued availability of datasets in external storage associated with accessing data processing systems.
It is another object of the present invention to provide an improved method and system for backup copying of records in external storage concurrent with a dramatically shortened suspension of data processing system application execution occasioned by such copying.
It is yet another object of the present invention to provide an improved method and system for incremental backup copying of records in a data processing system which minimizes, still further the suspension of data processing system application execution during such copying, as well as the actual amount of data which must be backed up.
The foregoing objects are achieved as is now described. Backup copying of designated datasets representing a first selected point in time consistency may be performed in a data processing system on an attached storage subsystem concurrent with data processing system application execution by first suspending application execution only long enough to form a logical-to-physical address concordance, and thereafter physically backing up the datasets on the storage subsystem on a scheduled or opportunistic basis. An indication of each update to a selected portion of the designated datasets which occurs after the first selected point in time is stored and application initiated updates to uncopied designated datasets are first buffered. Thereafter, sidefiles are made of the affected datasets, or portions thereof, the updates are then written through to the storage subsyste
Eastridge Lawrence Elwood
Kern Robert Frederic
Ratliff James Mitchell
Bracewell & Patterson L.L.P.
De'cady Albert
International Business Machines - Corporation
Schecter Manny W.
Ton David
LandOfFree
Method and system for incremental time zero backup copying... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and system for incremental time zero backup copying..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for incremental time zero backup copying... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2875184