The PASM prototype interconnection network design
نویسنده
چکیده
Computer system design costs traditionally have been dominated by the development and production costs of hardware. The availability of large-scale integrated circuitry has dramatically cut the cost of system hardware to the point where new architectural designs can be built and tested at a comparatively low cost. The P ASM prototype currently under development will be used to validate the design principles and operations of a large-scale, dynamically reconfigurable parallel processing sys. tern. The interconnection network of the PASM prototype, an implementation of a circuit-switched extra stage cube topology, is described in this paper. Network design tradeoffs and implementation options are discussed and related to the performance, development cost, and production cost of the network and the overall prototype system.
منابع مشابه
Fault Side-Effects in Fault Tolerant
Many fault tolerant (FT) multistage interconnection networks have been devised that tolerate one or more faults. It is shown that the effects of some realistic faults can propagate through multiple stages and cause non-faulty components to behave erroneously; such faults with side effects can cause "FT" networks to lose their fault tolerance capabilities even due to a single fault. The PASM pro...
متن کاملPERFORMANCE AND IMPLEMENTATION OF 4x4 SWITCHING NODES IN AN INTERCONNECTION NETWORK FOR PASM
Design issues for the multistage Generalized Cube network are discussed in this paper. An analysis of the merits of 2-input/2-output interchange boxes versus 4-input/4-output crossbars for interconnection network implementation is made. The cost and performance of each network for the two switching node alternatives are examined. Discussion of the sui tabi l i ty of each approach for VLSI imple...
متن کاملDesign and Analysis of Dynamic Redundancy Networks
Most previous work in the fault-tolerant design of multistage interconnection networks (MIN’s) has been based on improving the reliabilities of the networks themselves. For parallel systems containing a large number of processing elements (PE’s), the capability to recover from a PE fault is also important. The dynamic redundancy (DR) network is investigated in this paper. It can tolerate faults...
متن کاملPerformance Analysis of a New Neural Network for Routing in Mesh Interconnection Networks
Routing is one of the basic parts of a message passing multiprocessor system. The routing procedure has a great impact on the efficiency of a system. Neural algorithms that are currently in use for computer networks require a large number of neurons. If a specific topology of a multiprocessor network is considered, the number of neurons can be reduced. In this paper a new recurrent neural ne...
متن کاملPerformance Analysis of a New Neural Network for Routing in Mesh Interconnection Networks
Routing is one of the basic parts of a message passing multiprocessor system. The routing procedure has a great impact on the efficiency of a system. Neural algorithms that are currently in use for computer networks require a large number of neurons. If a specific topology of a multiprocessor network is considered, the number of neurons can be reduced. In this paper a new recurrent neural ne...
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 2010