https://ir.unisa.ac.za/handle/10500/32197 2026-05-05T13:11:59Z https://ir.unisa.ac.za/handle/10500/32236 Proceedings of the 5th/Vth Southern African Computer Symposium, Holiday Inn, Sandton, 29 November - 1 December 1989 (Full book) Kritzinger, P. (Ed.) Conference proceedings 1989-11-29T00:00:00Z https://ir.unisa.ac.za/handle/10500/32234 Integrated network management Roos, J.D.; Von Backström, L. Over the last decade, networks have evolved into sophisticated, multi-vendor, multi-protocol networks, which need to be properly managed and protected. The preferred way of accomplishing this, is through Integrated Network Management (INM). This paper deals with the concepts and advantages of INM, and addresses the trends in INM, particularly the role of Automated and Open Systems Interconnection (OSI) Network Management (NM). Performance evaluation 1989-11-29T00:00:00Z https://ir.unisa.ac.za/handle/10500/32233 A protocol engineering workstation - with an application to protocol performance analysis Kritzinger, P.S.; Wheeler, G. This paper describes the Protocol Engineering Workstation (PEW), an integrated specification, debugging and testing package based upon the Estelle FDT. It consists of three primary integrated components: a screen editor, an Estelle compiler, and an interpreter/debugger for the execution of Estelle specifications, as well as a separate analysis system. Apart from a brief description of each component which make up the package, it is shown how the PEW can be used with the analysis system to analyse the performance of the protocol being specified. The underlying model upon which the analysis is based, is outlined and a sample protocol specification and analysis illustrates the practical application of the system. Performance evaluation 1989-11-29T00:00:00Z https://ir.unisa.ac.za/handle/10500/32231 Predicting the performance of shared multiprocessor caches Goosen, H.A.; Cheriton, D.R. We investigate the performance of shared caches in a shared-memory multiprocessor executing parallel programs, and formulate simple models for estimating the load placed on the bus by such a shared cache. We analyze three parallel program traces to quantify the amount of sharing that takes place during program execution. These results indicate that shared caches can substantially reduce the load placed on a bus by a large number of processors. Finally we propose a new metric of group locality to aid in predicting the performance of a parallel program running on a machine with shared caches. Performance evaluation 1989-11-29T00:00:00Z