Bit Weaving: A Non-Prefix Approach to Compressing Packet Classifiers in TCAMs

Abstract

Abstract-Ternary Content Addressable Memories (TCAMs) have become the de facto standard in industry for fast packet classification. Unfortunately, TCAMs have limitations of small capacity, high power consumption, high heat generation, and high cost. The well-known range expansion problem exacerbates these limitations as each classifier rule typically has to be converted to multiple TCAM rules. One method for coping with these limitations is to use compression schemes to reduce the number of TCAM rules required to represent a classifier. Unfortunately, all existing compression schemes only produce prefix classifiers. Thus, they all miss the compression opportunities created by non-prefix ternary classifiers.In this paper, we propose bit weaving, the first non-prefix compression scheme. Bit weaving is based on the observation that TCAM entries that have the same decision and whose predicates differ by only one bit can be merged into one entry by replacing the bit in question with *. Bit weaving consists of two new techniques, bit swapping and bit merging, to first identify and then merge such rules together. The key advantages of bit weaving are that it runs fast, it is effective, and it is composable with other TCAM optimization methods as a pre/post-processing routine.We implemented bit weaving and conducted experiments on both real-world and synthetic packet classifiers. Our experimental results show the following: (i) bit weaving is an effective stand-alone compression technique (it achieves an average compression ratio of 23.6%) and (ii) bit weaving finds compression opportunities that other methods miss. Specifically, bit weaving improves the prior TCAM optimization techniques of TCAM Razor and Topological Transformation by an average of 12.8% and 36.5%, respectively.

Wildcard Compression of Inter-Domain Routing Tables for OpenFlow-Based Software-Defined Networking

“…Wildcard-based compression for ACLs is considered in [21]. It uses a heuristic that runs in polynomial time which is based on two functions: bit swapping and bit merging.…”

An Adaptable Rule Placement for Software-Defined Networks

“…Further, there are three paths A, B, and C traversing a switch. All three paths have these two rules in their ingress port policies, one of which is a permit rule r 1 with matching field of src: 10 Figure 5. If we decide to share the three rules, there will be a circular dependency since we have to put r 1 before r 2 for path A and B but the order is reversed for path C; otherwise, the semantics of the policies will not be obeyed.…”

“…These range from rule compression systems on single switches [7], [8], [9], [10], [11], [12], distributing ACL policies on network edge switches [13], distributing ACL policies while also changing routing on internal switches [14], as well as distributing ACL policies while respecting a separate routing module [1], [15]. Most approaches follow the idea of representing ACL policies using multi-dimensional packet (or flow) spaces, computing covers in some multi-dimensional packet space that corresponds to optimized rules, and distributing such covers over network paths.…”

Split: Optimizing Space, Power, and Throughput for TCAM-Based Classification

“…This work has used one of two types of techniques: equivalent transformation [2,6,13,14,16,21] and reencoding [4,5,9,15,17,19,25]. In equivalent transformation, the goal is to convert a given d-dimensional classifier into a semantically equivalent d-dimensional classifier with fewer TCAM entries.…”