Browse Prior Art Database

Cost Oriented Algorithm for Data Set Allocation in Storage Hierarchies Based on Fundamental Data Set Characteristics

IP.com Disclosure Number: IPCOM000082204D
Original Publication Date: 1974-Oct-01
Included in the Prior Art Database: 2005-Feb-28
Document File: 2 page(s) / 28K

Publishing Venue

IBM

Related People

Ling, H: AUTHOR [+4]

Abstract

This description is directed to allocating data sets to levels of a storage hierarchy on the basis of their activity, application characteristics and computer system and storage costs, including considerations such as movement of data sets from one level to another (staging and data migration). Open data sets may be allocated short of core on staging and closed data sets may be allocated short of archive on migration. Costs considered include storage cost, open cost, access cost, hardware cost, and data movement cost.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 2

Cost Oriented Algorithm for Data Set Allocation in Storage Hierarchies Based on Fundamental Data Set Characteristics

This description is directed to allocating data sets to levels of a storage hierarchy on the basis of their activity, application characteristics and computer system and storage costs, including considerations such as movement of data sets from one level to another (staging and data migration). Open data sets may be allocated short of core on staging and closed data sets may be allocated short of archive on migration. Costs considered include storage cost, open cost, access cost, hardware cost, and data movement cost.

When data sets are stored in a multilevel hierarchical storage system, it must be determined which data set stays at what level when inactive and where it will be moved to when it is opened. Data migration refers to moving data to archive storage when not used, and staging refers to moving data to higher speed device when opened.

A program written for data set allocation must consider a number of factors. It must decide where the data set is to be stored at the time of activity and inactivity. This means storage cost at both periods must be taken into account. If the two places of storage are not the same, it must consider the cost involved in moving the data set.

Further, when the data set becomes active it will be accessed in different ways, i.e., sequential or random, and the program must analyze this aspect. Also, the speed of the...