How to Emulate Fine-grained Multithreading
نویسندگان
چکیده
Fine-grained multithreading can be used to hide longlatency operations encountered in parallel computers during remote memory access. Instead of using special processor hardware, the emulation of fine-grained multithreading on standard processor hardware is investigated. While emulation of coarse-grained multithreading is common in modern operating systems, in the fine-grained case research on emulation has been limited and design of multithreaded processors has been favored. It will be shown that latencies encountered in todays parallel computers such as Cray T3E can be hidden by emulation of fine-grained multithreading using a moderate number of threads. Thus, emulation of fine-grained multithreading can be a viable alternative to the expensive design of custom processor hardware with support for multiple threads.
منابع مشابه
Grids and Clusters with Multi-Core Nodes: A Genetics Application Perspective
The introduction of multicore processors imply that algorithms which are parallelized at an outer, coarse grain level should possibly be revisited to examine if multithreading should also be used at an inner, fine grain level. In this paper we discuss parallel versions of the tightly coupled global optimization algorithm DIRECT. We examine how both coarse grained and fine grained parallelism ca...
متن کاملPredictable CPU Architecture Designed for Small Real-Time Application - Concept and Theory of Operation
The purpose of this paper is to describe an predictable CPU architecture, based on the five stage pipeline assembly line and a hardware scheduler engine. We aim at developing a fine-grained multithreading implementation, named nMPRA-MT. The new proposed architecture uses replication and remapping techniques for the program counter, the register file, and the pipeline registers and is implemente...
متن کاملFine-Grained Multithreading for the Multifrontal QR Factorization of Sparse Matrices
The advent of multicore processors represents a disruptive event in the history of computer science as conventional parallel programming paradigms are proving incapable of fully exploiting their potential for concurrent computations. The need for different or new programming models clearly arises from recent studies which identify fine-granularity and dynamic execution as the keys to achieve hi...
متن کاملThe Cray MTA and Unstructured Meshes
The Cray MTA, a multithreaded architecture, is a new parallel supercomputer installed at San Diego Supercomputer Center (SDSC). This machine has an architecture quite different from those of other contemporary parallel machines. It has a flat, shared memory without locality and has hardware support for very fine-grained multithreading. The machine and its associated parallelizing compiler promi...
متن کامل