Optimal Representations of a Traffic Distribution in Switch Memories

Sadeh, Yaniv; Rottenstreich, Ori; Barkan, Arye; Kanizo, Yossi; Kaplan, Haim

doi:10.1109/tnet.2020.2977477

Cited by 9 publications

(24 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper we assume the Longest Prefix Match (LPM) model in which wildcards appear only as a consecutive suffix of the pattern. This model is motivated by specialized TCAM hardware such as [15], and has been studied intensively [14,20,21,25]. As detailed in [11], common programmable switch architectures such as the RMT and Intel's FlexPipe have tables of different types and in particular tables dedicated to longest prefix matching [4,18].…”

Section: Traffic Splitting Problemmentioning

confidence: 99%

“…In this paper, we consider only partitions to parts which are non-negative integers that sum to 2 . The following natural problem has been solved in [21]. Problem 1.…”

Section: Traffic Splitting Problemmentioning

confidence: 99%

“…Focusing on the Longest Prefix Match model, [14] suggested an algorithm named Niagara, showed that it is very efficient in practice, and considered a tradeoff of reduced accuracy for less rules. [21] suggested an optimal algorithm which they called Bit Matcher that computes a minimal size TCAM for a desired partition, and proved that Niagara is also optimal.…”

mentioning

confidence: 99%

“…The work of [21] does not address the common scenario in which the available number of TCAM entries is smaller than the minimum needed to represent the desired partition exactly and therefore an approximate solution is necessary. However [21] showed experimentally that a truncation of an implementation of the exact partition gives a good approximation. Specifically, they showed that the subsets of the 'less specific' rules produced by Bit Matcher and Niagara, provide a good approximate partition according to several metrics.…”

mentioning

confidence: 99%

“…(1) We give new polynomial time algorithms that find partitions that best approximate a desired partition and can be represented with a given memory constraint. To the best of our knowledge, this problem has never been studied, except for simple heuristics in [14,21]. We do so for three "approximation-measures" between partitions.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Optimal approximations for traffic distribution in bounded switch memories

Sadeh

Rottenstreich

Kaplan

2020

Proceedings of the 16th International Conference on Emerging Networking EXperiments and Technologies

Self Cite

View full text Add to dashboard Cite

Traffic splitting is a required functionality in networks, for example for load balancing over multiple paths or among different servers. The capacities of the servers determine the partition by which traffic should be split. A recent approach implements traffic splitting within the ternary content addressable memory (TCAM), which is often available in switches. It is important to reduce the amount of memory allocated for this task since TCAMs are power consuming and are often also required for other tasks such as classification and routing. Previous work showed how to compute the smallest TCAM necessary to implement a given partition exactly. In this paper we solve the more practical case, where at most TCAM rules are available, restricting the ability to implement the desired partition. We give simple and efficient algorithms to find rules that generate a partition closest in ∞ to the desired one. We do the same for a one-sided version of ∞ which equals to the maximum overload on a server and for a relative version of it. We use our algorithms to evaluate how the expected error changes as a function of the number of rules, the number of servers, and the width of the TCAM.

show abstract

Section: Traffic Splitting Problemmentioning

confidence: 99%

“…In this paper, we consider only partitions to parts which are non-negative integers that sum to 2 . The following natural problem has been solved in [21]. Problem 1.…”

Section: Traffic Splitting Problemmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations