Design of a Data Storage Hierarchy. DSH-III--Software & Hardware.

Abstract

Modern organizations are becoming increasingly reliant on the storage and processing of very large data bases in support of their accounting, operational control, and high-level decision-making functions. Contemporary Data Base Management Systems are capable of handling data bases on the order of a trillion bits of data, and can process transactions at rates of up to 100 queries per second. However, it is expected that future high-performance DBM's will be required to provide storage capacities and transaction rates several orders of magnitude greater than those of any current systems. It is not unreasonable to project requirements of a quadrillion bits and one million database accesses per second. This report presents a refined set of control algorithms which take advantage of theoretical properties of INFOPLEX-like storage systems. Section 2 presents an overview of the INFOPLEX design issues and tradeoffs related to performance and reliability, and develops a general strategy for efficient READ/WRITE control and reliable fault handling. Section 3 develops the functional design of DSH-III in more detail, including functional descriptions of the system components and their interfaces. Sections 4 and 5 detail the algorithms and protocols supporting the READ and WRITE operations, respectively. Section 6 summarizes this report, and indicates the directions for future research.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Mar 01, 1986
Accession Number
ADA166287

Entities

People

  • Michael J. Abraham

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Access Time
  • Algorithms
  • Application Software
  • Automatic
  • Computers
  • Data Management
  • Data Storage Systems
  • Data Transmission
  • Databases
  • Decomposition
  • Mass Storage
  • Operating Systems
  • Parallel Computing
  • Parallel Processing
  • Reliability
  • Software Design
  • User Interface

Fields of Study

  • Computer science

Readers

  • Business Analytics
  • Database Systems and Applications
  • Parallel and Distributed Computing.