Multi-processor systems are becoming more commonplace and affordable. Based on analyses of actual parsings, we argue that to exploit the capabilities of such machines, unification-based grammar parsers should distribute work at the level of individual unification operations. We present a generic approach to parallel chart parsing that meets this requirement, and show that an implementation of this technique for LinGO achieves considerable speedups.

In terms of both speed and memory consumption, graph unification remains the most expensive component of unification-based grammar parsing. We present a technique to reduce the memory usage of unification algorithms considerably, without increasing execution times. Also, the proposed algorithm is thread-safe, providing an efficient algorithm for parallel processing as well.