Dennis C. Guster

Learn More
Distributed processing has provided a number of advantages to network designers. These service advantages include better performance, backup and availability potential. This paper uses simulation to predict network performance of three host models: single host, dual hosts and dual hosts with mirrored file systems. The results indicate that the dual models(More)
The paper provides an overview of three common virtualization threats that have been observed in log files in the authors' network and suggests solutions to mitigate those security vulnerabilities. The solutions offered have been implemented on a network with over 200 hosts 40 of which are virtualized.
In an organization, a well devised disaster recovery plan is not only crucial in the information recovery process, but also vital in the quest to sustain daily operations. While prior research has discussed many recovery sites options, assessment of recovery site communication paths and their associated complexity is still limited in regard to the(More)
The advent of distributed processing and global information systems has driven the need for fast, efficient and secure global authentication systems. Typically distributed processing implies that any given application will require computing resources from multiple computing nodes, and hence for the sake of convenience will require single sign-on capability(More)
The advent of distributed processing has drastically increased the number of hosts required to support an enterprise level computing domain. Often, each application such as accounting, DNS or WWW service is allocated its own dedicated host for performance and security reasons. While this scenario is effective from an enterprise computing perspective, it(More)
Data mining and very large databases offer a great deal of promise for business and government alike. Unfortunately, due to the ever changing data needs of the modern organization, as evidenced by the growing number of petabyte data applications, the task to efficiently extract data in a timely manner becomes more difficult. The research described herein(More)