Answers are grounded in this paper and cite their sources.
Recently, we proposed a two-stage pipelined Bloom filter architecture to save power for network security applications. In this letter, we generalize the pipelined Bloom filter architecture to k-stage and show that significant power savings can be achieved by employing one hash function per stage. We analytically show that the expected power consumption and latency of the fully pipelined Bloom filter architecture will not be greater than that of the two hash functions and two clock cycles, respectively, however large the number of hash functions is. Furthermore, we discuss the worst-case performance of the proposed architecture.
No takes yet. Share an insight, caveat, or question.
Paynter et al. (2008) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: