Secure Distributed Processing Systems.
Abstract
In chapter 2 we identify the major issues in the design of distributed systems, namely: interconnection structures, error recovery, user interface, security and program and data assignment. Also, in this chapter we discuss some of the aspects of the design of distributed database management systems such as data integrity, concurrency control, crash recovery and synchronization protocols. Chapter 3 addresses the issue of concurrency control and deadlock handling in distributed databases. A locking protocol with adaptive centralized control and distributed recovery procedures is presented. A hierarchical extension of this protocol is also proposed. In the same chapter, we present two deadlock detection protocols for distributed database - a distributed one and a hierarchically organized one. Chapter 4 is concerned with issues of crash recovery in computer systems. In particular, it addresses the following problem. Given a set of objects (e.g. processes, files, etc.), a set of snapshots for each of them and the records of information flow between them, find the set of objects and their snapshots which should be used to restore the state of the system to a consistent state, once an error in any of the (potentially interacting) objects is detected. An efficient algorithm to perform consistent state restoration in a centralized computer system is given in this chapter. Protocols for performing consistent state restoration in distributed systems are developed and presented in chapter 5. Finally, chapter 6 presents and outline of the architecture of the UCLA Distributed Secure System Base.
Document Details
- Document Type
- Technical Report
- Publication Date
- Dec 01, 1978
- Accession Number
- ADA186683
Entities
People
- Gerald J. Popek
Organizations
- University of California, Los Angeles