Efficient multi-field packet classification for QoS purposes

Abstract

Mechanisms for service diferentiation in datagram networks, such as the Internet, rely on packet classification in routers to provide appropriate service. Classification involves matching multiple packet header fields against a possibly large set of filters identihing the diferent service classes. In this paper, we describe a packet classifier based on tries and binomial trees and we investigate its scaling properties in three QoS scenarios that are likely to occur in the Internet. One scenario is based on Integrated Sem'ces and RSVP and the other two are based on Differentiated Sewices. By performing a series of tests, we characterize the processing and memory requirements for a software implementation of our classifier. Evaluation is done using real data sets takenfrom two existing high speed networks. Resultsfrom the IntServ/RSVP tests on a Pentium 200MHz show that it takes about 10.5 p per packet and requires 2,000 KBytes of memory to classify among 22,000 entries. Classification for a virtual leased line service based on DifServ with the same number of entries takes about 9 p per packet and uses less than 250 KBytes of memory. With an average packet size of 2,000 bits, our classifier can manage data rates of about 200 Mbps on a 200 MHz Pentium. We conclude that multi-feld classification is feasible in software and that high performance classifiers can run on low cost hardware. IntroductionConnectionless datagram networks rely on perpacket classification in routers. Traditional best-effort unicast packet forwarding is done by classifymg destination addresses of packets against a set of address prefixes. This is known as single-field classification. The increasing use of the Internet for business and commercial purposes introduces an economic incentive for providing service differentiation. Also, increasing availability and popularity of applications with real-time constraints help drive the evolution of service differentiation.To support service differentiation between packets, classification must be extended to involve also 'Computer Science and Electrical Engineering LuleA University of Technology SE-971 87 Luled Sweden olov@cdt.luth.seother header fields. This is known as multi-field classification. There are several application areas for multi-field packet classification (e.g., service differentiation, firewalls, QoS routing, etc.). In this paper we present a QoS classifier and we evaluate it specifically in the context of providing service differentiation in IP networks.To provide service differentiation, the IETF has standardized Integrated Services (IntServ) [12] and the Resource Reservation protocol (RSVP) [15] and is currently working on a more scalable solution called Differentiated Services (DiffServ) [1][2]. The DiffServ effort was motivated, among other things, by the scaling problems of the IntServ/RSVP model resulting from the need for per-flow state in routers. A basic idea with DiffServ is that complexity is pushed to the edges of the network to keep the network core free from per-flow state...

Bilateral Teleoperation System Using QoS and Secure Communication Networks for Telemedicine Applications

“…However, applied to the network-based bilateral teleoperation system, IPSec needs some enhancements. In fact, in both QoS standardized architectures (IntServ [33] and DiffServ [34]), the multifield (MF) packet classifier inspects incoming packets and classifies them by setting a priority in the type-of-service field [35]. This way, bilateral transmission packets and commands will normally have higher priorities than any other TCP/UDP normal traffic and, thus, will be assigned appropriate QoS parameters (bandwidth, delay, jitter, and packet loss).…”

Specification, implementation and performance evaluation of the QoS‐friendly encapsulating security payload (Q‐ESP) protocol

“…As the classifier inspects multiple fields in the packet, it is called Multi-Field (MF) classifier [7].…”

Design of multi-field IPv6 packet classifiers using ternary CAMs

“…Some of these papers presented the software designs [4], [6], [7] to find out the matched rule, and the others [8], [9], [13] designed hardware architectures, for example, using CAMs, to accelerate the searching procedure. However, in the worst case, these algorithms may suffer from the problem of backtracking in the searching procedure, which is undesired in the ultra high-speed lookup engines.…”

“…The famous trie structures have been proposed in several articles to compare the five fields of IP/TCP headers sequentially [1], [2], [7], [8]. Also multi-dimension structures have been proposed for the same purpose [10], [12].…”