PROJECT TITLE :

Encoding Short Ranges in TCAM Without Expansion: Efficient Algorithm and Applications - 2018

ABSTRACT:

We gift range encoding with no enlargement (RENÉ)- a completely unique encoding theme for brief ranges on Ternary content addressable memory (TCAM), that, in contrast to previous solutions, will not impose row growth, and uses bits proportionally to the maximal vary length. We have a tendency to provide theoretical analysis to point out that our encoding is the closest to the lower certain of range of bits used. Yet, we have a tendency to show many applications of our technique in the sector of packet classification, and additionally, how the same technique might be used to efficiently solve different exhausting problems, like the closest-neighbor search downside and its variants. We have a tendency to show that using TCAM, one might solve such issues in abundant higher rates than previously instructed solutions, and outperform known lower bounds in traditional memory models. We tend to show by experiments that the interpretation method of RENÉ on switch hardware induces solely a negligible two.five% latency overhead. Our nearest neighbor implementation on a TCAM device provides search rates that are up to four orders of magnitude above previous best previous-art solutions.