Abstract

With the advent of multi-core processing in personal computing systems, the demand for concurrent software has increased far beyond its original niche. The conventional model of lock-based concurrency for writing such software is difficult in use, error-prone and does not always lead to efficient use of the available resources. In this paper we shall evaluate several alternative models, namely the Actor Model, the Join-Calculus, Software Transactional Memory and Runtime Scheduling and compare them to the conventional one, to determine whether they provide superior means of writing concurrent software.

Additional Resources

1. The paper