• ETRI Journal
• /
• v.26 no.5
• /
• pp.505-508
• /
• 2004

We propose a new arbitration method for an input buffered switch with a buffered crossbar. In the proposed method, an exhaustive polling method is used to decrease the synchronization. Using an approximate analysis, we explain how the proposed method improves the switch performance. Also, using computer simulations, we show the proposed method outperforms the previous methods under burst traffic.

• The KIPS Transactions:PartC
• /
• v.10C no.4
• /
• pp.485-492
• /
• 2003

This paper presents a new modified input buffered switch, which called a grouped input buffered (GIB) switch, to eliminate the influence of HOL blocking when using multiple input buffers in ATM switches. The GIB switch consists of grouped sub switches per a network stage. The switch gives extra paths and buffered switching elements between groups for transferring the blocked cells. As the result, the proposed model can reduce the effect by the HOL blocking and thereafter it enhances the performance of the switch. The simulation results show that the proposed scheme has good performance in comparison with previous works by using the parameters such as throughput, cell loss, delay and system power.

• Proceedings of the IEEK Conference
• /
• 2000.11a
• /
• pp.283-286
• /
• 2000

One major drawback of conventional output buffered switches is that the speed of writing cells into output buffer should be N times faster than input link speed. This paper proposes a new output buffer switch that divides one output buffer into several buffers and virtually shares the divided buffers by using a distributor. The proposed switch makes it possible to reduce the memory speed. The proposed switch is evaluated in terms of the average cell latency compared with the input buffered switches which use the arbitration alogorithms, i.e., iSLIP or wrapped wave front arbiter(WWFA).

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 1997.11a
• /
• pp.147-161
• /
• 1997

In the future, the performance of B-lSDN offering the multimedia and a various service depends on the performance of switch that is the important factor consisting of network. Bufferless banyan network consisted of MIN(multistage interconnection network) selected for- the fabric of ATM switch and has a limitation of performance because of blocking. Input buffered banyan networks with FIFO(first-in first-out) buffering scheme for the reduction of blocking and the cell bypass queueing theory for the reduction of HOL(head of line) blocking were seperately compared of the performance of switch. Specially input buffered banyan networks were applied monotonic buffering scheme that was proposed. As a result of simulation, Buffered Banyan Network with cell bypass queueing theory showed better performance than FIFO type input buffered Banyan network. Monotonic increase buffering scheme showed better performance than Monotonic decrease buffering scheme.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.1A
• /
• pp.114-121
• /
• 2000

The problem of an input-buffered switch is the HOL(head-of-line) blocking which limits the maximum throughput but it is easy to implement in hardware. However, HOL blocking can be eliminated using aVOQ(virtual-output-queueing) technique. 0 this paper, we propose a new cell-scheduling algorithm for aninput-buffered ATM switch. The proposed algorithm, called PPIM(Probabilistic Parallel Iterative Matching), imposesa weight to every request based on the reserved bandwidth. It is shown that the input-buffered ATM switch withthe proposed PPIM algorithm not only provides high throughput and low delay but it also reduces the jitter,compared with the existing WPIM(Weighted PIM).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.2 no.1
• /
• pp.117-124
• /
• 1998

As window sized increases, the throughput input-buffered packet switch with a window scheme improves on random traffic condition. However, the improvement diminishes quickly under bursty traffic. In this paper, we propose Burst Head Addressable FIFO mechanism and memory structure having search capability in unit of burst header to compensate the sensitiveness of the windowed scheme to bursty traffic. The performance of a input-buffered switch using the proposed Burst Header Addressable FIFO input buffer was analyzed using computer simulations. The maximum throughput of the conventional FIFO scheme approaches an asymptotic value 0.5 as mean burst length increases. The maximum throughput of the proposed scheme is greater than that of the conventional scheme for any mean burst length and window size.

• Korean Management Science Review
• /
• v.10 no.1
• /
• pp.139-156
• /
• 1993

Various ATM switches have been proposed since Asynchronous Transfer Mode (ATM) was recognized as appropriate for implementing broadband integrated services digital network (BISDN). An ATM switching network may be evaluated on two sides : traffic performances (maximum throughput, delay, and packet loss probability, etc.) and structural features (complexity, i.e. the number of switch elements necessary to construct the same size switching network, maintenance, modularity, and fault tolerance, etc.). ATM switching networks proposed to date tend to show the contrary characteristics between structural features and traffic performance. The Knockout Switch, which is well known as one of ATM switches, shows a good traffic performance but it needs so many switch elements and buffers. In this paper, we propose an input and output buffered Knockout Switch for the purpose of reducing the number of switch elements and buffers of the existing Knockout Switch. We analyze the traffic performance and the structural features of the proposed switching architecture through a discrete time Markov chain and compare with those of the existing Knockout Switch. It was found that the proposed architecture could reduce more than 40 percent of switch elements and more than 30 percent of buffers under a given requirement of cell loss probability of the switch.

• The KIPS Transactions:PartC
• /
• v.10C no.4
• /
• pp.463-470
• /
• 2003

Many types of switching fabrics have been proposed for use in ATM networks. Multistage Interconnection Networks (MINs) constitute a large class of ATM switching systems that are widely used in today´s internetworking. One of the most veil-known types of multistage networks is the banyan network. The banyan network is attractive for its simple routing scheme and low hardware complexity, but its throughput is very limited due to internal blocking and output contention. In this paper, we propose an input-buffered dual-banyan switch model with multiple switching fabric between switch input and output to avoid internal and Head-of Line blocking. By performance analysis and simulation, we show that our model has a lower ceil delay and 96% throughput which is much better than other banyan-type switch architecture.

• Journal of Electrical Engineering and information Science
• /
• v.3 no.1
• /
• pp.28-35
• /
• 1998

In this paper, an NxN input-buffered crosspoint packet switch which selects a Head of the Line, HOL, packet in contention randomly is analyzed with a new approach. The approach is based on both a Markov chain representation of the input buffer and the probability that a HOL packet is successfully served. The probability as a function of N is derived, and it makes it possible to express the average packet delay and the average number of packets in the buffer as a function of N. The contention resolution policy based on the occupancy of the input buffer is also presented and analyzed with this same approach and the relationship between two selection policies is analyzed in terms of the occupancy of the input buffer.

• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.117-120
• /
• 1999

Input queueing is useful for high bandwidth switches and routers because of lower complexity and fewer circuits than output queueing. The input queueing switch, however, suffers HOL-Blocking, which limits the throughput to 58%. To get around this low throughput, we propose a simple scheduling algorithm called Synchronous Input Port (SIP). This method synchronize packets and switching without blocking, which is shown to have better performance over the established algorithms