Reviewer: Max Podlesny
Date: 11-17-2005
How would you rate this paper, relative to others we have read? top 25%, but not top 10%
How would you rate your kowledge of the topic of this paper? familiar, but not expert
What problem or issue does the paper address? Why is it important?
The problem considered in the paper is of designing a mechanism tghat would allow to perform multi-match packet classification. The problem of packet classification is actual for intrusion detection systems, packet-level accounting systems. The main idea of the proposed mechanism is to split filters into groups and perform filtering in each group separately. As groups of filters are not correlated it allows to parallelize lookups into filters and update them fast.
What are the main contributions of the paper and why are they important?
Th main contribution of the paper is of developing a Set Splitting Algorithm classifying packets that reduces th TCAM memory and power consumption.
How significant are these contributions relative to previous work?
Th main siginificance of the contributions is reducing the TCAM memory and power consumption.
Give detailed comments justifying your view of the paper.
The paper consideres an important problem and good results are obtained. The structure is good, all important issues are considered in the paper. I did not find any substancial weakness in the paper.