Method and means for cataloging data sets using dual keyed data sets and direct pointers
First Claim
1. A method for operating a computing system under control of a catalog management facility referencing a catalog data set residing on a first data volume to open a user data set residing on a second data volume, where said first or second data volume may have been changed by events external to the catalog management facility, resulting in loss of synchronization between said user data set and said catalog data set, comprising the steps of;
- searching said catalog data set for a first record keyed to said user data set, said first record including a pointer to a second record in a volume data set residing on said second data volume;
responsive to said pointer, retrieving said second record;
testing the key of said second record to detect a loss of synchronization; and
,if said second record is keyed to said user data set and thus indicates synchronization, opening the user data set described by said second record;
otherwise, searching said volume data set in said second data volume to retrieve a third record keyed to said user data set, and upon locating said third record, updating said pointer to reestablish synchronization between said pointer and said user data set, and opening the user data set described by said third record;
whereby said user data set is opened without signaling an error to the user even though loss of synchronization has previously resulted from an external event such as a restoration of said first data volume or said second data volume from an archival data volume.
0 Assignments
0 Petitions
Accused Products
Abstract
A data set catalog structure eliminates the requirement for base catalog/data volume synchronization in a multi-processing environment while enabling the operating efficiency directly addressing the data volumes. The catalog is distributed between a keyed sequential base catalog and, on each data volume, an entry sequential volume data set. Catalog information which must be synchronized with application data sets is stored in volume records in the volume data set.
A method for opening a user'"'"'s data set having a key comprises the steps of obtaining from the base catalog a direct pointer associated with the key to the volume record for the user data set; accessing the volume record (relative byte address) addressed by the pointer; comparing the key of the user data set with the key of the accessed volume record; if the keys do not compare, key searching the volume data set to locate the volume record containing correct key and rewriting the direct pointer in the base catalog to address the correct volume record, and obtaining from the correct volume record a direct pointer to the user data set.
-
Citations
21 Claims
-
1. A method for operating a computing system under control of a catalog management facility referencing a catalog data set residing on a first data volume to open a user data set residing on a second data volume, where said first or second data volume may have been changed by events external to the catalog management facility, resulting in loss of synchronization between said user data set and said catalog data set, comprising the steps of;
-
searching said catalog data set for a first record keyed to said user data set, said first record including a pointer to a second record in a volume data set residing on said second data volume; responsive to said pointer, retrieving said second record; testing the key of said second record to detect a loss of synchronization; and
,if said second record is keyed to said user data set and thus indicates synchronization, opening the user data set described by said second record;
otherwise, searching said volume data set in said second data volume to retrieve a third record keyed to said user data set, and upon locating said third record, updating said pointer to reestablish synchronization between said pointer and said user data set, and opening the user data set described by said third record;whereby said user data set is opened without signaling an error to the user even though loss of synchronization has previously resulted from an external event such as a restoration of said first data volume or said second data volume from an archival data volume. - View Dependent Claims (2, 3, 4, 5)
-
-
6. A method for operating a computing system under control of a catalog management facility referencing a catalog data set, the computing system including at least one central processing unit coupled to at least one data volume, to open a user data set in said data volume, the user data set being identified by a data set key, and wherein said data volume may have been changed by events unknown to said catalog management facility, resulting in loss of synchronization between said user data set and said catalog data set, the method comprising the steps of:
-
(1) retrieving from said catalog data set a sphere record associated by key with said user data set; (2) fetching from said sphere record a first address to a first volume record in a keyed volume data set residing on said data volume; (3) responsive to said first address, retrieving from said volume data set said first volume record; (4) comparing the key of said first volume record with said data set key to detect a loss of synchronization; (5) if said keys do not compare equal, key searching said volume data set for the volume record containing the correct key, and updating said sphere record to contain the address of the correct volume record; and (6) building user data set control blocks from data set characteristics contained in the volume record key associated with said user data set; whereby said user data set is opened without signaling an error to a user even though the catalog data set and the user data set may have lost synchronization as a result of an external event such as restoration of said data volume from an archival data volume. - View Dependent Claims (7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A method for operating a computing system to catalog user data sets on a plurality of data volumes, the computing system including at least two loosely coupled processors sharing access to at least one of said data volumes, the method establishing a catalog structure, including a base catalog structure, residing on the shared data volume, which enables the opening of a user data set on a second data volume even where an external event has changed a data volume, resulting in loss of synchronization between said base catalog and said user data set, the method comprising the steps of:
-
defining a base catalog structure by; allocating space on said shared data volume for the base catalog structure; generating a data volume record and an index volume record for describing the base catalog structure data and index components; writing said data volume record and said index volume record into addressable locations in a volume data set on said shared data volume; inserting the catalog name and pointers to said data volume record and index volume record into a volume control record in said volume data set; generating a catalog sphere record including direct address pointers to said data volume record and said index volume record; and writing said catalog sphere record into said base catalog structure on said shared data volume; and defining a plurality of user data sets into a catalog, each data set having a key name, and the catalog including said base catalog structure and said volume data set, by executing the following steps for each user data set to be defined; allocating space on said second data volume for a user data set; generating one or more user volume records for describing said user data set; writing said user volume records into an addressable location in said volume data set; generating a user sphere record, including direct address pointers to said user volume records and the key names of the user data set components; writing said user sphere record into said base catalog structure. - View Dependent Claims (17)
-
-
18. Apparatus for cataloging user data sets stored on a data volume to enable opening of a user data set by a central processor having access to the data volume and a main storage, said apparatus enabling opening of a user data set even though residing on a volume which, due to some external event, has been changed to cause loss of synchronization, the apparatus comprising:
-
means providing a volume data set for storing a plurality of volume records each containing identification information including at least one key field and extent information for user data sets recorded on the data volume; base catalog means for storing a plurality of sphere records, each sphere record containing one or more key fields identifying a user data set and a direct address pointer to a corresponding volume record; means for opening a first user data set having a key name by building, in the main storage, control blocks providing user data set identification and extent information, said means for opening including; first means for searching said base catalog means for a first sphere record containing said key name in its key field; second means responsive to the direct address pointer stored in said first sphere record for retrieving a first volume record; third means for comparing said key name with a first key field of said first volume record; fourth means, responsive to a failure of correspondence between said key name and said first key field and thus indicative of a loss of synchronization, for searching said volume data set means for a second volume record containing a second key field corresponding to said key name, and for writing into said first sphere record a direct address pointer to said second volume record; and means for retrieving information defining the extent of said first user data set from that one of said volume records which contains a key field corresponding to said key name. - View Dependent Claims (19, 20, 21)
-
Specification