Throughput computing with chip multithreading and clusters

Authors:
Mukund Buddhikot;Sanjay Goil
Affiliations:
Sun Microsystems, Santa Clara, CA;Sun Microsystems, Santa Clara, CA
Venue:
HiPC'05 Proceedings of the 12th international conference on High Performance Computing
Year:
2005

Citing 4
Cited 0

Simultaneous multithreading: maximizing on-chip parallelism

ISCA '95 Proceedings of the 22nd annual international symposium on Computer architecture
Design and implementation of the POWER5™ microprocessor

Proceedings of the 41st annual Design Automation Conference
Niagara: A 32-Way Multithreaded Sparc Processor

IEEE Micro
High-Performance Throughput Computing

IEEE Micro

Quantified Score

Hi-index	0.00

Visualization

Abstract

Chip MultiThreading (CMT) based systems are being introduced in the market by several computer platform vendors. At the same time, cluster computing platforms are becoming prevalent in market segments which tend to be highly price/performance driven. This paper analyzes the architectural space of these two prominent computing paradigms in technical computing markets. This analysis is carried out in terms of the application turnaround time, throughput, and scalability across multiple threads of execution. Additionally, we introduce various subscription models to optimize application throughput and turnaround time.