Technical Report Number
Due to the importance and complexity of the packet classiﬁcation problem, a myriad of algorithms and re-sulting implementations exist. The performance and capacity of many algorithms and classiﬁcation devices, including TCAMs, depend upon properties of the ﬁlter set and query patterns. Unlike microprocessors in the ﬁeld of computer architecture, there are no standard performance evaluation tools or techniques avail-able to evaluate packet classiﬁcation algorithms and products. Network service providers are reluctant to distribute copies of real ﬁlter sets for security and conﬁdentiality reasons, hence realistic test vectors are a scarce commodity. The small subset of the research community who obtain real ﬁlter sets either limit performance evaluation to the small sample space or employ ad hoc methods of modifying those ﬁlter sets. In response to this problem, we present ClassBench, a suite of tools for benchmarking packet classiﬁcation algorithms and devices. ClassBench includes a Filter Set Generator that produces synthetic ﬁlter sets that accurately model the characteristics of real ﬁlter sets. Along with varying the size of the ﬁlter sets, we provide high-level control over the composition of the ﬁlters in the resulting ﬁlter set. The tools suite also includes a Trace Generator that produces a sequence of packet headers to exercise the synthetic ﬁlter set. Along with specifying the relative size of the trace, we provide a simple mechanism for controlling locality of reference in the trace. While we have already found ClassBench to be very useful in our own research, we seek to initiate a broader discussion and solicit input from the community to guide the reﬁnement of the tools and codiﬁcation of a formal benchmarking methodology.
Taylor, David E. and Turner, Jonathan S., "ClassBench: A Packet Classification Benchmark" Report Number: WUCSE-2004-28 (2004). All Computer Science and Engineering Research.