Electrical computers and digital processing systems: memory – Storage accessing and control – Specific memory composition
Reexamination Certificate
1999-05-18
2001-10-16
Kim, Matthew (Department: 2186)
Electrical computers and digital processing systems: memory
Storage accessing and control
Specific memory composition
C711S154000, C711S170000, C711S202000
Reexamination Certificate
active
06304940
ABSTRACT:
TECHNICAL FIELD
This invention relates to the sharing of data sets between incompatible data processor systems. More particularly, the invention relates to a direct access storage system for accomplishing the sharing of data sets between two incompatible data processor systems.
BACKGROUND OF THE INVENTION
The difficulty of making Multiple Virtual Storage (MVS) files available to other systems is well known in the art. There are two aspects to making count, key and data (CKD) format files from a MVS available to a system which employs Fixed Block Architecture files: getting the data from one system to another in a form allowing sharing, and recoding it from one system's conventions to the other.
The files may be transferred, rather than shared, by using a simple text transfer, for example, by using an internet binary image mode, which is stream binary. Each internet connection sets up a separate connection and this is extremely inefficient and error prone. The form of transfer does not accomplish sharing of the data sets.
A method for transferring large data sets to another system is to write a tape and move the tape to the other system for reading, using one tape drive that is compatible with the MVS system and another tape drive that can read the same tape, but which is compatible with the other system. The data sets on each system after the transfer are totally separate and are not shared.
Utilities may be able to edit the data to convert text files, but are handled very inefficiently and transferred as a single file, and the transfer may not be error free.
Data mining is the ability to analyze a data file to interpret the file and ascertain information that is not apparent from a direct reading of the file. Examples include analyzing credit card data to ascertain the most likely group of persons who spend money more than a particular amount by credit card a month based on deposit and age, and to then select a more detailed group who have deposits, for example, between $3000 and $4000 and are in their 40's.
Another example is to ascertain unusual credit card activity and thereby identify possible credit card fraud.
In order to perform an accurate analysis, error free data is required.
In order to provide an effective analysis to, for example, identify a compromised credit card promptly, the access to the data set must be prompt.
Lastly, the data set to be analyzed may be changing rapidly so that it would be advantageous to provide continuing updates to the file as it is being analyzed.
Most credit card systems that would be analyzed are MVS systems. Most data mining applications are done by data processors running UNIX or AIX which employ FBA data, which data is incompatible with the CKD data of the MVS systems. The typical data set to be analyzed is in the range of 10's or 100's of gigabytes, which would require many hours of transmission time between systems using communication techniques.
Several techniques have been developed for emulating CKD on FBA but typically are usable for only small data sets or a limited number of records, and/or do not allow access by an FBA device, and therefore do not allow sharing of the data. Examples are U.S. Pat. No. 5,206,939 to Yanai et al. which discloses a compression technique for representing the COUNT field of every record in a cascading arrangement, U.S. Pat. No. 5,301,304 to Menon which discloses an emulation program which completely emulates CKD data stored on FBA devices while still retaining the same byte displacements of the CKD data by elimination of padding, intra-record gaps, etc., U.S. Pat. No. 5,283,884 to Menon et al. which stores a table entry in non-volatile memory for each record, and U.S. Pat. No. 5,535,372 to Benhase et al. which provides a track format descriptor in non-volatile storage for each track and describes whether the track comprises “well behaved” formatted data which is simple to search or is not “well behaved” such that the entire track must be loaded into cache.
It may be desirable to transfer the results of the data mining back to the originating data processing system to identify the specific accounts meeting the criteria of the analysis so that specific mailings may be conducted or phone calls placed, etc. Thus, a user requirement may be that access to the data set must be maintained by the originating system for proper coordination.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a data storage system and method which allow data sharing on a high speed direct access basis between an MVS system and a data processing system employing FBA data. Hereinafter, a data processing system employing CKD data is referred to either as an MVS data processing system, or a CKD data processing system.
Disclosed are a shared direct access storage system and sharing method for use with both an MVS data processing system and a data processing system employing FBA data, for storing data set records received from the MVS data processing system. The MVS data processing system may extract the data set from an on line data base. A CKD control element creates an MVS Volume Table of Contents (VTOC) for the data set from the MVS data processing system and stores the data set in CKD data storage. An FBA controller and disk data storage provides for direct access to FBA data by the FBA data processing system. A mapping element, which may be a program, logic, or processor, maps the data set records from the MVS data processing system into shadow volume data records with fixed blocks. The received data set records are uniformly sized, and the mapping element converts the uniformly sized records into smaller sized fixed blocks, with the size of the uniformly sized records being an even multiple of the smaller sized fixed blocks. A mapping algorithm translates the CKD volume id, track id and record number for the first record of the data set as identified in the VTOC, of the shadow volume into the starting offset FBA volume id for the first block and the id's for the continuing sequence of fixed blocks. The data fields of the shadow volume records are mapped into FBA data records of fixed blocks, with each FBA data record arranged in a linear sequence of fixed blocks, as stored in the CKD disk data storage for direct access by the FBA controller for the data processor employing FBA data.
The VTOC and the address of the sequence of fixed blocks may be established by the shared data storage system, or a series of extents may be defined for the MVS host and utilized by the host to establish the VTOC, and the mapping algorithm establishes the logical FBA addresses.
The data set remains a CKD shadow volume data set, under the control of the CKD control element and of the MVS host, while being shared by the FBA controller. The MVS data processing system may thus provide updates to the data set.
The FBA mapping volume is identified to the FBA controller and its host as either a “raw disk”, or, preferably, as a “removable media”.
For a fuller understanding of the present invention, reference should be made to the following detailed description taken in conjunction with the accompanying drawings.
REFERENCES:
patent: 5206939 (1993-04-01), Yanai et al.
patent: 5257362 (1993-10-01), Menon
patent: 5283884 (1994-02-01), Menon et al.
patent: 5301304 (1994-04-01), Menon
patent: 5388013 (1995-02-01), Nakamura
patent: 5459853 (1995-10-01), Best et al.
patent: 5463754 (1995-10-01), Beausoliel et al.
patent: 5535372 (1996-07-01), Benhase et al.
patent: 5564019 (1996-10-01), Menon
patent: 5581743 (1996-12-01), Burton et al.
patent: 5636359 (1997-06-01), Beardsley et al.
patent: 5664144 (1997-09-01), Yanai et al.
patent: 5724542 (1998-03-01), Taroda et al.
patent: 5860088 (1999-01-01), Benhase et al.
patent: 5862363 (1999-01-01), Taroda et al.
patent: 5920893 (1999-07-01), Nakayama et al.
patent: 6009498 (1999-12-01), Kumasawa et al.
patent: 6041386 (2000-03-01), Bello
patent: 6112277 (2000-08-01), Bui et al.
patent: 6115797 (2000-09-01), Kanda et al.
Clio, D. Blokkum, et al., Proceedings-Share Europ
Bataille Pierre-Michel
Holcombe John H.
International Business Machines - Corporation
Kim Matthew
Sullivan Robert M.
LandOfFree
Shared direct access storage system for MVS and FBA processors does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Shared direct access storage system for MVS and FBA processors, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Shared direct access storage system for MVS and FBA processors will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2613318