A Retrieval Strategy for Large, Multi-Key Files Requiring Frequent Updating,
Abstract
Techniques are developed which have practical application in multi-key retrieval operations with large files. The concepts of block addressing, related to hardware characteristics, and data transformations, to reduce the volume of descriptors, are introduced. Application of these principles results in a process for interrogating files ranging in size from hundreds to millions of items. Moreover, the system provides efficient, on-line file maintenance. An important criterion for direct access of large files is the amount of extra storage required to facilitate access. The overhead requirements for the process described in this paper will normally represent less than five percent. The overhead can be adjusted higher or lower to reflect desired increases or decreases in retrieval efficiency on a field or file basis. The techniques are independent of actual data values; however, an improvement in response may be generated based on usage of the file. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1971
- Accession Number
- AD0890465
Entities
People
- Edgar M. Cagley