Recovery techniques are an important aspect of database systems. They are essential to ensure that data integrity is maintained after any type of failure occurs. The recovery mechanism must be designed so that the availability and performance of the system are not unacceptably impacted by the recovery algorithms running during normal execution. On the other hand, enough information must be stored so that the database can be restored or transactions backed out in a reasonable amount of time. Concepts, techniques, and problems associated with database recovery will be presented in this thesis. The recovery issues for both centralized and distributed systems will be discussed, along with the tradeoffs of different recovery tools. The database recovery schemes in IMS/VS, DB2 and SDD-1 will be described to show approaches in existing systems.

Library of Congress Subject Headings

Database management; Database management--Evaluation

Publication Date


Document Type


Student Type


Degree Name

Computer Science (MS)

Department, Program, or Center

Computer Science (GCCIS)


Henry Etlinger

Advisor/Committee Member

Peter Anderson

Advisor/Committee Member

Jeffrey Lasky


Physical copy available from RIT's Wallace Library at QA76.9.D3E42 1987


RIT – Main Campus