Research on Interconnection Network for Massively Parallel Machines

نویسنده

  • Shigeru Oyanagi
چکیده

Interconnection network is one of the most important problems in designing massively parallel machines. Interconnection network must be designed to balance the processing speed and communication speed. This problem becomes more serious when the scale of the machine is huge. Our research aims to propose mechanisms for interconnection network that is expandable toward million processor systems. At the former half of this project, our research will be based on the electronic implementation, and at the latter half, it will be extended to optical interconnection.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

RAMA: a file system for massively-parallel computers

This paper describes a file system design for massively parallel computers which makes very efficient use of a few disks per processor. This overcomes the traditional I/O bottleneck of massively parallel machines by storing the data on disks within the high-speed interconnection network. In addition, the file system, called RAMA, requires little inter-node synchronization, removing another comm...

متن کامل

A Hypercube-based Scalable Interconnection Network for Massively Parallel Computing

An important issues in the design of interconnection networks for massively parallel computers is scalability. A new scalable interconnection network topology, called Double-Loop Hypercube (DLH), is proposed. The DLH network combines the positive features of the hypercube topology, such as small diameter, high connectivity, symmetry and simple routing, and the scalability and constant node degr...

متن کامل

Strategies for the Implementation of Interconnection Network Simulators on Parallel Computers

Methods for simulating multistage interconnection networks using massively parallel SIMD computers are presented. Aspects of parallel simulation of interconnection networks are discussed and different strategies of mapping the architecture of the network to be simulated onto the parallel machine are studied and compared. To apply these methods to a wide variety of network topologies, the discus...

متن کامل

A Strategy for Efficient and Scalable Collective Communication in the Quadrics Network

Large-scale parallel machines are traditionally designed to provide a computing platform for many scientific problems, that require a huge amount of computing power. Parallel scientific applications span many different fields such as climate modeling, computational fluid dynamics and protein folding. During the past decade a common strategy to build high-performance parallel machines has called...

متن کامل

Network Performance under Physical Constraints

The performance of an interconnection network in a massively parallel architecture is subject to physical constraints whose impact needs to be re-evaluated from time to time. Fat-trees and low dimensional cubes have raised a great interest in the scientiic community in the last few years and are emerging standards in the design of interconnection networks for massively parallel computers. In th...

متن کامل

ذخیره 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 1994