Technology & Research

Intel® Technology Journal Home

Volume 11, Issue 03

Tera-scale Computing


Intel Technology Journal - Featuring Intel's recent research and development

ISSN 1535-864X DOI 10.1535/itj.1103.01

  • Volume 11
  • Issue 03
  • Published August 22, 2007

Tera-scale Computing

Section 1 of 10  

Integration Challenges and Tradeoffs for Tera-scale Architectures

Mani Azimi, Corporate Technology Group, Intel Corporation
Naveen Cherukuri, Corporate Technology Group, Intel Corporation
D. N. Jayasimha, Corporate Technology Group, Intel Corporation
Akhilesh Kumar, Corporate Technology Group, Intel Corporation
Partha Kundu, Corporate Technology Group, Intel Corporation
Seungjoon Park, Corporate Technology Group, Intel Corporation
Ioannis Schoinas, Corporate Technology Group, Intel Corporation
Aniruddha S. Vaidya, Corporate Technology Group, Intel Corporation

Index words: tiled architecture, on-die interconnect, cache hierarchy, communication protocol

Citations for this paper. Azimi, M.; Cherukuri, N.; Jayasimha, D.N.; Kumar, A.; Kundu, P.; Park, S.; Schoinas, I.; Vaidya, A. "Integration Challenges and Tradeoffs for Tera-scale Architectures." Intel Technology Journal. http://www.intel.com/technology/itj/2007/
v11i3/1-integration/1-abstract.htm
(August 2007).

ABSTRACT

Tera-scale processors promise to offer an unprecedented concentration of computing power and enable novel usages and applications. The computing power may be provided by a combination of general-purpose cores and special-purpose (fixed or programmable) computing engines. Further, Moore's law enables the integration of additional system resources to the processor die. However, the realization of tera-scale architecture is challenged by on-die power dissipation, wire delays, off-chip memory bandwidth, process variations, and higher failure rates. These challenges create opportunities for architectural innovation. One of the ways to address these challenges is through the use of a "tiled" architecture: the die is divided into a large number of identical or close-to-identical, tiles that are interconnected using a scalable and energy-efficient interconnect. This modular approach enables ease of layout and rapid integration of different blocks. Limited off-chip memory bandwidth requires innovations in the cache hierarchy, memory subsystem, and coherence protocol. We present an architectural vision for the tera-scale processors and discuss the performance, scalability, and manufacturability aspects of the uncore. We articulate key challenges and point to candidate solutions for these challenges.

Section 1 of 10  

Back to Top

In This Article

Download a PDF of this article.