Dimitrios Stiliadis

High-speed policy-based packet forwarding using efficient multi-dimensional range matching

The ability to provide differentiated services to users with widely varying requirements is becoming increasingly important, and Internet Service Providers would like to provide these differentiated services using the same shared network infrastructure. The key mechanism, that enables differentiation in a connectionless network, is the packet classification function that parses the headers of the packets, and after determining their context, classifies them based on administrative policies or real-time reservation decisions. Packet classification, however, is a complex operation that can become the bottleneck in routers that try to support gigabit link capacities. Hence, many proposals for differentiated services only require classification at lower speed edge routers and also avoid classification based on multiple fields in the packet header even if it might be advantageous to service providers. In this paper, we present new packet classification schemes that, with a worst-case and traffic-independent performance metric, can classify packets, by checking amongst a few thousand filtering rules, at rates of a million packets per second using range matches on more than 4 packet header fields. For a special case of classification in two dimensions, we present an algorithm that can handle more than 128K rules at these speeds in a traffic independent manner. We emphasize worst-case performance over average case performance because providing differentiated services requires intelligent queueing and scheduling of packets that precludes any significant queueing before the differentiating step (i.e., before packet classification). The presented filtering or classification schemes can be used to classify packets for security policy enforcement, applying resource management decisions, flow identification for RSVP reservations, multicast look-ups, and for source-destination and policy based routing. The scalability and performance of the algorithms have been demonstrated by implementation and testing in a prototype system.

Latency-rate servers: a general model for analysis of traffic scheduling algorithms

We develop a general model, called latency-rate servers (/spl Lscr//spl Rscr/ servers), for the analysis of traffic scheduling algorithms in broadband packet networks. The behavior of an /spl Lscr//spl Rscr/ server is determined by two parameters-the latency and the allocated rate. Several well-known scheduling algorithms, such as weighted fair queueing, virtualclock, self-clocked fair queueing, weighted round robin, and deficit round robin, belong to the class of /spl Lscr//spl Rscr/ servers. We derive tight upper bounds on the end-to-end delay, internal burstiness, and buffer requirements of individual sessions in an arbitrary network of /spl Lscr//spl Rscr/ servers in terms of the latencies of the individual schedulers in the network, when the session traffic is shaped by a token bucket. The theory of /spl Lscr//spl Rscr/ servers enables computation of tight upper bounds on end-to-end delay and buffer requirements in a heterogeneous network, where individual servers may support different scheduling architectures and under different traffic models.

Efficient fair queueing algorithms for packet-switched networks

Although weighted fair queueing (WFQ) has been regarded as an ideal scheduling algorithm in terms of its combined delay bound and proportional fairness properties, its asymptotic time complexity increases linearly with the number of sessions serviced by the scheduler, thus limiting its use in high-speed networks. An algorithm that combines the delay and fairness bounds of WFQ with O(1) timestamp computations had remained elusive so far. In this paper we present two novel scheduling algorithms that have O(1) complexity for timestamp computations and provide the same bounds on end-to-end delay and buffer requirements as those of WFQ. The first algorithm, frame-based fair queueing (FFQ), uses a framing mechanism to periodically recalibrate a global variable tracking the progress of work in the system, limiting any short-term unfairness to within a frame period. The second algorithm, starting potential based fair queueing (SPFQ), performs the recalibration at packet boundaries, resulting in improved fairness while still maintaining the O(1) timestamp computations. Both algorithms are based on the general framework of rate-proportional servers (RPSs) introduced by Stiliadis and Varma (see ibid., vol.6, no.2, p.164-74, 1998). The algorithms may be used in both general packet networks with variable packet sizes and in asynchronous transfer mode (ATM) networks.

Rate-proportional servers: a design methodology for fair queueing algorithms

Generalized processor sharing (GPS) has been considered as an ideal scheduling discipline based on its end-to-end delay bounds and fairness properties. Until recently, emulation of GPS in a packet server has been regarded as the ideal means of designing a packet-level scheduling algorithm to obtain low delay bounds and bounded unfairness. Strict emulation of GPS, as required in the weighted fair queueing (WFQ) scheduler, however, incurs a time-complexity of O(N) where N is the number of sessions sharing the link. Efforts in the past to simplify the implementation of WFQ, such as self-clocked fair queueing (SCFQ), have resulted in degrading its isolation properties, thus affecting the delay bound. We present a methodology for the design of scheduling algorithms that provide the same end-to-end delay bound as that of WFQ and bounded unfairness without the complexity of GPS emulation. The resulting class of algorithms, called rate-proportional servers (RPSs), are based on isolating scheduler properties that give rise to ideal delay and fairness behavior. Network designers can use this methodology to construct efficient fair-queueing algorithms, balancing their fairness with implementation complexity.

Relative differentiated services in the Internet: issues and mechanisms

In the context of relative differentiated services the network traffic is grouped in N classes of service which differ in their relative forwarding quality, i.e., class i is better than class i 1 in terms of local (per-hop) forwarding measures of the queueing delays and packet losses. Users and applications, on the other hand, select adaptively the class that best meets their requirements and/or cost constraints. In this short paper we present, focusing on queueing delays only, four different service models that offer such a class relative differentiation. Simulation results illustrate that a predictable and controllable delay differentiation can be achieved in only one of these models. A more complete treatment of this subject appears in [I].

Performance bounds of rate-adaptation schemes for energy-efficient routers

To maximize the energy efficiency of packet networks, energy use in network equipment should scale rigorously with the traffic load. Rate adaptation is gaining popularity as a promising framework for achieving energy proportionality in the data-path components of routers and switches, but a thorough characterization of its energy-saving capabilities and impact on network performance is still lacking. In this paper, we present novel rate-adaptation schemes that are easy to implement, amenable to incremental deployment in packet switches and routers, and whose negative effect on end-to-end packet delays is bounded. We derive tight lower and upper bounds on energy consumption that we use to further refine the design of our schemes. Our analysis indicates that large-scale deployments of rate adaptation can enable substantial energy savings, as large gains can be attained in individual data-path devices under parametric configurations that do not compromise the end-to-end delay performance of the network.

Architecture of an integrated router interconnected spectrally (IRIS)

The design of optical packet routers poses significant challenges both in terms of its architecture and component design. In this paper, we evaluate several alternatives for the architecture of such routers, and describe the architecture of IRIS (integrated router interconnected spectrally), an optical router being designed at Bell Laboratories. By combining load balancing with wavelength switching, the IRIS architecture can make use of thousands of wavelengths and provide terabits of capacity, well above the scalability limits of router architectures based on other approaches. The IRIS architecture uses load balancing to eliminate the need for centralized scheduling, and wavelength switching to allow N2 channels in an NtimesN space switch. We describe several architectural schemes for overcoming the limitations of the underlying optical devices in the design of IRIS. We also present two methods to improve the utilization of the optical buffers in IRIS to achieve high performance even with a small number of buffers

Demonstration of a time buffer for an all-optical packet router

We demonstrate an optical time buffer technology that meets the requirements of an all-optical packet router with a load-balancing architecture. The buffer is based on wavelength switching and an arrayed waveguide grating, which selects one of the delays in an array of fiber delay lines of increasing length. This buffer is unique because it enables the maximal sharing of delay lines among multiple input ports. Each of the N delays can simultaneously support N different wavelengths; therefore, for N input ports, the total buffer capacity is of the order of N3 packets. Measurements performed with data at 10 Gbits/s and N=3 show that there is only a 2 dB bit error rate (BER) penalty resulting from wavelength switching.

Efficient multicast algorithms for high-speed routers

Several types of applications, like broadcast video, interactive gaming, remote collaboration, etc., require multicast distribution of content. An important problem, from the router architecture perspective, is to distribute multicast content at high-speeds and over a large number of physical or logical interfaces. In order to forward multicast packets, the system must replicate the same packet, or pointers of the packet, to multiple interfaces. This function requires significant speed-up from the memories and the forwarding engines, and has become a key bottleneck for wide deployment of multicasting. Previous efforts to solve this problem have only led to statistical solutions that cannot provide a deterministic performance. As a result, temporary overload in multicast traffic could lead to unnecessary packet losses, even for unicast connections that share the same interfaces. We present a novel algorithm for managing multicast sessions, that offers isolation between interfaces, deterministic performance, and flexibility in the offered delays. To the best of our knowledge, this is the first algorithm proposed that can provide deterministic performance for multicast sessions in high-speed switches and routers, without any speed-up requirements.

Energy-efficient data transfer primitives for laptops using mobile handhelds

We introduce a novel mechanism for content distribution to large numbers of weakly connected laptops that can be switched off frequently and have intermittent network access. Relying on the users data-enabled mobile phone and a gateway added to the data path between the laptop and the Internet, the mechanism builds upon two novel data-transfer primitives that efficiently move files across the network even when the laptop is switched off or sleeping, in a way that is fully transparent to the application layer. One primitive targets network-folder-based applications, while the other works for web-based applications. The primitives have been successfully deployed in the field as part of a solution for remote IT management of mobile-employee laptops.