Interconnection Networks for Large - Scale Parallel Computers
نویسندگان
چکیده
A new scalable interconnection topology called the Spanning Multichannel Linked Hypercube (SMLH) is proposed along with an optical implementation methodology that combines both the advantages of free space optics with those of wavelength division multiplexing techniques. The SMLH uses the hypercube topology as a basic building block and connects such building blocks using two-dimensional multichannel links (similar to spanning buses). In doing so, the SMLH combines positive features of both the hypercube (small diameter, high connectivity, symmetry, simple routing, and fault tolerance) and the spanning bus hypercube (SBH) (constant node degree, scalability, and ease of physical implementation), while at the same time circumventing their disadvantages. The SMLH topology supports many communication patterns found in di erent classes of computation, such as bus-based, mesh-based, and treebased problems as well as hypercube-based problems. A very attractive feature of the SMLH network is its ability to support a large number of processors with the possibility of maintaining a constant degree and a constant diameter. Other positive features include symmetry, incremental scalability, and faulttolerance. It is shown that the SMLH network provides better average message distance, average tra c density, and queueing delay than many similar networks, including the binary hypercube, the SBH, etc. Additionally, the SMLH has comparable performance to other high-performance hypercubic networks, including the Generalized Hypercube and the Hypermesh. This work was supported by the U.S. National Science Foundation under grant number MIP 93-10082. 1
منابع مشابه
On the Theory of Interconnection Networks for Parallel Computers
EEcient data transfer between processors is an essential component in any large scale parallel computation. Motivated by the growing interest in parallel computers, a signiicant amount of theoretical research has been devoted to the area of interconnection networks for parallel computers , most of it to the packet routing (or store-and-forward) model of communication. We survey some of the majo...
متن کاملDual-cubes: a New Interconnection Network for High-performance Computer Clusters
The binary hypercube, or n-cube, has been widely used as the interconnection network in parallel computers. However, the major drawback of the hypercube is the increase in the number of communication links for each node with the increase in the total number of nodes in the system. This paper introduces a new interconnection network for large-scale distributed memory multiprocessors called dual-...
متن کاملA Distributed Management Scheme for Partitionable Parallel Computers
Many large-scale parallel computers, such as those based on hypercube or multistage cube interconnection networks, can be partitioned to process applications with different computation structures and different degrees of parallelism simultaneously. However, partitioning may create resource fragments and may result in a loss of computation power. Dynamic Partitioning is an effective method to al...
متن کاملAn E cient 3 D Optical Implementation of Binary de Bruijn Networks with Applications to Massively Parallel Computing 1
The interconnection network structure can be the deciding and limiting factor in cost and performance of parallel computers. One of the most popular point-to-point interconnection networks for parallel computers today is the hypercube. The regularity, logarithmic diameter, symmetry, high connectivity, fault-tolerance, simple routing, and recon gurability (easy embedding of other network topolog...
متن کاملNSIM: An Interconnection Network Simulator for Extreme-Scale Parallel Computers
In the near future, interconnection networks of massively parallel computer systems will connect more than a hundred thousands of computing nodes. The performance evaluation of the interconnection networks can provide real insights to help the development of efficient communication library. Hence, to evaluate the performance of such interconnection networks, simulation tools capable of modeling...
متن کاملWormhole Routing in Parallel Computers
In order to offer low communication latency and reduced buffer requirements, wormhole routing has been used in almost all new generation parallel computers. It has been a powerful switching technique where its communication latency is distance insensitive. On the other hand, deadlock avoidance is the most critical issue in wormhole networks. If deadlock handling is not considered, wormhole rout...
متن کامل