-
[show abstract]
[hide abstract]
ABSTRACT: Buffered Clos-network switch, also referred as memory-memory-memory (MMM) Clos-network switch, is an alternative to single-stage switches to implement large-scale packet switches. In this paper, we unveil the head-of-line blocking problem of MMM Clos-network switches with switch modules implemented by buffered crossbars, which causes performance degradation. We propose a three-stage buffered Clos-network switch with per-output flow queues in the switch modules at the first two stages to avoid head-of-line blocking. The proposed switch, called the MM �� M switch, achieves higher performance than an MMM buffered Clos-network packet switch. We show the performance improvement of the proposed MM �� M switch under different traffic patterns.
01/2011;
-
ICTAI 2009, 21st IEEE International Conference on Tools with Artificial Intelligence, Newark, New Jersey, USA, 2-4 November 2009; 01/2009
-
[show abstract]
[hide abstract]
ABSTRACT: The growth of the Internet and requirements for enhanced flexibility and versatility have resulted in packet classification becoming an essential part of network systems such as routers and firewalls. Classification of packets into flows based on multiple header fields is a complex problem. Several algorithms have been proposed to achieve wire speed classification performance. However, an ever widening gap between wire speed and memory access speed continues to motivate the quest for techniques to enhance classification performance. In this paper, we propose a caching technique to improve search performance of any classification algorithm using characteristics inherent to Internet traffic.
Sarnoff Symposium, 2008 IEEE; 05/2008
-
Proceedings of IEEE International Conference on Communications, ICC 2008, Beijing, China, 19-23 May 2008; 01/2008
-
[show abstract]
[hide abstract]
ABSTRACT: Current schemes for configuration of input-queued three-stage Clos-network (IQC) switches involve port matching and path routing assignment, in that order. The implementation of a scheduler capable of matching thousands of ports in large-size switches is complex. To decrease the scheduler complexity for such switches (e.g., 1024 ports or more), we propose a configuration scheme for IQC switches that hierarchizes the matching process. In a practical scenario our scheme performs routing first and port matching thereafter. This approach reduces the scheduler size and the configuration complexity of IQC switches. We show that the switching performance of the proposed approach using weight-based and weightless selection schemes is high under uniform and nonuniform traffic
IEEE Communications Letters 03/2007; · 0.98 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: This chapter presents an overview of internally buffered crossbar switches as a feasible and attractive alternative to high-performance
packet switches for the next generation networks. This type of switch requires that memory works at the same speed as that
for IB switches to provide high switching performance. Most internally buffered switches have been shown to outperform IB
switches and to emulate OB switches with a speedup no more than 2. Therefore, internally buffered switches result in an economic
and efficient solution to the implementation of packet switches. Issues remaining for future research include the provisioning
of guaranteed services using internally buffered crossbar switches and the resolution of scalability issues. As line rates
continue to increase and memory speedup falls behind, new architectures with more memory-efficient strategies will be needed.
06/2006: pages 121-146;
-
[show abstract]
[hide abstract]
ABSTRACT: Stability of a switch or router is one of the features of most interest. A stable input buffered switch means that occupancies in any of the input queues converges to a value different from infinity for admissible ingress traffic. Techniques to study stability in a switch are limited and sometimes unsuitable for elaborated schemes. Lyapunov functions have been used to analyze stability in maximum weight matching (MWM) schemes and may be hard to apply in other schemes. Here, we propose an alternative technique and it may be applied to a wide range of selection schemes, and furthermore, to different switch architectures. As an application of this technique, this paper discusses the impact of the granularity of the service time in the stability of an input-queued switch.
Sarnoff Symposium, 2006 IEEE; 04/2006
-
[show abstract]
[hide abstract]
ABSTRACT: Networks with quality of service (QoS) provisioning rely on QoS routing schemes to select paths between end-to-end hosts that can satisfy the requested service requirements. In this paper, we focus on networks that use a new service paradigm as a nested DiffServ model, which has been described as service vectors in combination with the explicit end-point admission control (EEAC) scheme [1]. The main objectives of service vectors are to improve the QoS granularity and service routing flexibility. These improvements enhances both network utilization and user benefits. However, QoS routing schemes need to consider service vectors to achieve such improvements. In this paper, we discuss a new set of requirements that are added into QoS routing schemes to avoid false routing and low network utilization when using service vectors. The OSPF protocol, as the widely used protocol, guarantees feasible deployability of service vectors in existing networks. Furthermore, we introduce a network architecture that integrates security into the set of QoS parameters and show how security-enabled QoS (SQoS) can also use the OSPF protocol for SQoS routing.
Sarnoff Symposium, 2006 IEEE; 04/2006
-
[show abstract]
[hide abstract]
ABSTRACT: In the SQoS network as introduced in [1] and [2], the border router in every autonomous system (AS) provides customized security mechanisms to the incoming packets. Some serious problems have been recently raised particularly when there are one or more compromised routers that attempt to modify, delete, or fabricate any part or the whole packet into the SQoS network. The compromised router can either passively or actively perform the malicious activities against the forwarding packets. The SQoS network does not explicitly specify the method to detect whether the data contained in the packets have been tampered by the compromised routers or by the end host itself. We deliberate the threats and propose several methods to detect both the malicious routers and end hosts such that SQoS information and payload are authentic and integrity-protected.
Sarnoff Symposium, 2006 IEEE; 04/2006
-
IEICE Transactions. 01/2006; 89-B:1495-1504.
-
Proceedings of the Global Telecommunications Conference, 2006. GLOBECOM '06, San Francisco, CA, USA, 27 November - 1 December 2006; 01/2006
-
IEEE Communications Letters. 01/2005; 9:661-663.
-
IEEE Communications Letters. 01/2004; 8:585-587.
-
[show abstract]
[hide abstract]
ABSTRACT: Combined input-crosspoint buffered (CICB) packet switches have been of research interest in the last few years because of their high performance. These switches provide higher performance than input-buffered (IB) packet switches while requiring the crosspoint buffers run at the same speed as that of the input buffers in IB switches. Recently, it has been shown that CICB switches with one-cell crosspoint buffers, virtual output queues, and simple input and output arbitrations, provide 100% throughput under uniform traffic. However, it is of general interest to know the maximum throughput that a CICB switch, with no speedup, can provide under admissible traffic. This paper analyzes the throughput performance of a CICB switch beyond uniform traffic patterns and shows that a CICB switch with one-cell crosspoint buffers can provide 100% throughput under admissible traffic while using no speedup.
-
[show abstract]
[hide abstract]
ABSTRACT: Measurement is a required process in high performance networks for efficient quality-of-service (QoS) provisioning and service verification. Active measurement is an attractive approach because the measurement traffic injected into the network can be controlled and the measurement tasks can be distributed throughout the network. However, the execution of measurement tasks in common parts of a network may face contention for resources, such as computational power, memory, and link bandwidth. This contention could jeopardize measurement accuracy and affect network services. This contention for limited resources defines a conflict between measurement tasks. Furthermore, we consider two sets of measurement tasks, those used to monitor network state periodically, called periodic tasks, and those for casual measurements issued as needed, called on-demand measurement tasks. In this paper, we propose a novel scheduling scheme to resolve contention for resources of both periodic and on-demand measurement tasks from graph coloring perspective, called ascending order of the sum of clique number and degree of tasks. The scheme selects tasks according to the ascending order of the sum of clique number and conflict task degree in a conflict graph and allows concurrent execution of multiple measurement tasks for high resource utilization. The scheme decreases the average waiting time of all tasks in periodic measurement tasks scheduling. For on-demand measurement tasks, the proposed scheme minimizes the waiting time of inserted on-demand tasks while keeping time space utilization high. In other words, the total time spent on finishing all the tasks is shortened. We evaluate our proposed schemes under different measurement task assignment scenarios through computer simulations, and compare the performance of this scheme with others that also allow concurrent task execution. The simulation results show that the proposed scheme produces effective contention resolution and low execution delays.
Computer Communications.
-
[show abstract]
[hide abstract]
ABSTRACT: Single-stage input-queued (IQ) switches are attractive for implementation of high performance routers because they require no speedup in the used memory. It has been shown that IQ switches can provide 100% throughput under admissible traffic when using maximum-weight matching schemes or iterative maximal-weight matching schemes with a speedup of two or more. These different approaches require either high computation complexity or high memory costs that can make them infeasible. Therefore, there is a need for low-complexity and fast matching schemes that provide high throughput under several admissible traffic patterns, including those with nonuniform distributions, without recurring to speedup nor multiple iterations. In this paper, we introduce the concept of captured frame and apply it to matching schemes. As a result, we propose two weightless matching schemes, one based on round-robin selection, called uFORM, and the other based on random selection, called uFPIM. We analyze the throughput improvement achieved by uFPIM, and show that these matching schemes provide high throughput under a variety of admissible traffic patterns, including those with nonuniform distributions, when using a single iteration and no speedup. Furthermore, we study the scalability of the captured-frame concept in matching schemes for memory-space-memory multiple-stage Clos-network switches, and show the achieved high switching performance and low implementation complexity.
Computer Communications.
