International Business Machines Corporation

David Rolland Welsh

Thomas Paul Giordano

Thomas Owen McKinley

Barry Warren Knapp

Gary Ross Ricard

Robert J Bestgen

Larry William Youngren

John J Vriezen

Michael Lawrence Nordstrom

Robert Paul Mech

Mukesh Rathor

A method and computer program product are provided for processing and combining data sets including bitmaps in a computer system. A plurality of initial data subsets to be combined are processed and data structure statistics containing predefined data attributes are built for each of the plurality of initial subsets. Checking for a dynamic selection attribute for each of the plurality of initial subsets is performed. Responsive to the dynamic selection attribute not being identified, the intermediate storage requirements for producing the combined final data set is selectively optimized. Responsive to identifying the dynamic selection attribute, both intermediate result regeneration and intermediate storage requirements for producing a combined final data set are selectively optimized.

Method and computer program product for processing and combining data sets including bitmaps

A method and computer program product are provided for paging control using a reference structure in a computer system. The reference structure is scanned to identify a next selected entry for an IO range building routine. The next selected entry is compared with a set hardlimit value. Responsive to the next selected entry being greater than the hardlimit value, the IO range building routine is exited. A shortlimit value is identified. The next selected entry is compared with the identified shortlimit value. Responsive to the next selected entry being greater than the identified shortlimit value, the IO range building routine is exited. A first array is used for storing entry IDs for selected entries found from scanning the reference structure and a second array is used for tracking blocks of storage used for the selected entries.

Method and computer program product for paging control using a reference

In parallel process computer systems multiple tasks are simultaneously executed. However, where resource contention exists, the multiple tasks become serialized, reducing system performance. Once such resource contention is arm contention during reads for files stored under multiple arms on direct access storage devices (DASDs). According to the preferred embodiment, a method is provided that assigns read tasks in a manner that reduces the probability of arm contention during the file reads. First, the plurality of extents storing the file are correlated to their associated arm. Second, for each of the plurality of arms, the attributes of extents under that arm, preferably including the extent read times, are accumulated. Finally, the plurality of simultaneous read tasks are assigned to the plurality of extents according to the accumulated attributes such that arms that have a greater need for service are favored over arms that have a lesser need when reading the file. Assigning the tasks in this manner lessons the probability of arm contention during file reads, increasing system performance.

Computer mechanism for reducing DASD arm contention during parallel

The present invention involves a database administrator that is comprised of a plurality of database managers and a plurality of database servers. The database managers act as an interface between the database servers and client programs and the database servers act as an interface between the database managers and the database itself. The client programs, database managers, and database servers all execute within a single-level-store environment, which allows them to communicate with one another without incurring the expense of inter-address space communication and/or the expense associated with the cross-address space mapping of query results.

Parallel database serving mechanism for a single-level-store computer

An in use table manager in a computer system uses an in use table to track the use of files, or objects. The in use table is used to determine which objects may need recovery in the event of a system failure. Object addresses are hashed by the in use manager to identify a preferred slot in the table. The slots contain information identifying the object, and indicating the extent of use of the object. The in use manager assigns alternate slots, and dynamically changes the size of the in use table to reduce contention for slots. Several atomic operations on the table ensure integrity of the table, while permitting concurrent use. Portions of the table are bundled into single I/O operations to enhance system performance by minimizing I/O.

Growing community of inventors

Byron, MN, United States of America

David Rolland Welsh