A two-dimensional scalable crossbar matrix switch architecture
نویسندگان
چکیده
Rapid growth of Internet traffic causes a new challenge in the design of high-speed switches. One of main design issues for high-speed switches is a scalability problem. This paper proposes a scalable crossbar matrix (SCM) switch architecture, which consists of multiple crossbar switch units(XSU) with virtual output queues(VOQs) at the inputs and single-cell scheduling decomposition buffers(SDBs) at the outputs. We propose a distributed scheduling algorithm for the proposed scalable crossbar switch architecture, which consists of a credit based SLIP(C-SLIP) scheduling for crossbar switch units and backlog weighted round robin (BWRR) scheduling for switching fabric output ports. In this paper, we show that largescale switch fabric can be built up by small-size crossbar switch units with improved delay performance and enough arbitration time margin. The simulation results show that the proposed switch architecture and distributed scheduling algorithm can provide 100% throughput under i.i.d uniform traffic with a single iteration arbitration in a time slot. Keywords-crossbar switch, scalability, distributed scheduling.
منابع مشابه
A Class of Highly Scalable Optical Crossbar-Connected Interconnection Networks (SOCNs) for Parallel Computing Systems
ÐA class of highly scalable interconnect topologies called the Scalable Optical Crossbar-Connected Interconnection Networks (SOCNs) is proposed. This proposed class of networks combines the use of tunable Vertical Cavity Surface Emitting Lasers (VCSEL's), Wavelength Division Multiplexing (WDM) and a scalable, hierarchical network architecture to implement large-scale optical crossbar based netw...
متن کاملA Parallel-Polled Virtual Output Queued Switch with a Buffered Crossbar
* This material is based upon work supported by the National Science Foundation under Grant No. 9875177. Abstract Input buffered switches with Virtual Output Queues (VOQ) are scalable to very high speeds, but require switch matrix scheduling algorithms to achieve high throughput. Existing scheduling algorithms based on parallel requestgrant-accept cycles cannot natively support variable length ...
متن کاملNext Generation Internet High-Speed Switches and Routers
Shared memory architecture for packet switches was normally thought to be unsuitable for building high performance switches/routers. The main reason lies in their perceived poor scalability. In particular, shared memory architectures are typically used to build output-queued switches which are regarded as the best candidate to achieve optimal delay-throughput performance. The current trend in r...
متن کاملPerformance Evaluation of Variable Packet Size Buffered Crossbar Switches
The crossbar is very popular for the implementation of switches with a moderate number of ports. Unbuffered crossbars (IQ architecture) require complex switch matrix scheduling algorithms that operate on fixed-size cells. On the other hand, buffered crossbars (CICQ architecture), use small buffers at the crosspoints and provide excellent features; scheduling is simplified and variable-size pack...
متن کاملFast and Noniterative Scheduling in Input-Queued Switches
Most high-end switches use an input-queued or a combined inputand output-queued architecture. The switch fabrics of these architectures commonly use an iterative scheduling system such as iSLIP. Iterative schedulers are not very scalable and can be slow. We propose a new scheduling algorithm that finds a maximum matching of a modified I/O mapping graph in a single iteration (hence noniterative)...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2003