Download now Free registration required
Regular expression (Regex) becomes the standard signature language for security and application detection. Deterministic Finite Automata (DFAs) are widely used to perform multiple regex matching in linear time. However, when implemented by modern memories, the matching speed turns out to be a trade-off with the size of DFA. To improve the performance, the authors propose a generalized caching scheme that strikes the boundaries of memory size. They define the concept of local prediction which predicts the memory accesses to the DFA and guides the cache to be replaced with proper states so that the cache hit rate is greatly raised. The idea of using predictive DFA matching specifies an entire new class of approaches.
- Format: PDF
- Size: 360.5 KB