United States Home | Select a Location
Site Map | Contact Us | About Intel
Advanced Search
Home ComputingBusinessDeveloperReseller / Provider
R & D/technologies
Volume 6 Issue 1
Hyper-Threading Technology
Table of Contents
Reader's Guide
E-mail this journal to a colleague
Download PDF of this entire issue: Hyper-Threading Technology 1276k
Intel Technical Journal
By Lin Chao
Publisher, Intel Technology Journal

This February 2002 issue of the Intel Technology Journal (ITJ) is full of new things. First, there is a new look and design. This is the first big redesign since the inception of the ITJ on the Web in 1997. The new design, together with inclusion of the ISSN (International Standard Serial Number), makes it easier to index articles into technical indexes and search engines. There are new "subscribe", search ITJ, and "email to a colleague" features in the left navigation tool bar. Readers are encouraged to subscribe to the ITJ. The benefit is subscribers are notified by e-mail when a new issue is published.

The focus of this issue is Hyper-Threading Technology, a new microprocessor architecture technology. It makes a single processor look like two processors to the operating system. Intel's Hyper-Threading Technology delivers two logical processors that can execute different tasks simultaneously using shared hardware resources. Hyper-Threading Technology effectively looks like two processors on a chip. A chip with this technology will not equal the computing power of two processors; however, it will seem like two, as the performance boost is substantial. Chips enabled with Hyper-Threading Technology will also be cheaper than dual-processor computers: one heat sink, one fan, one cooling solution, and one chip are what are necessary.

The six papers in this issue of the Intel Technology Journal discuss this new technology. The papers cover a broad view of hyper-threading technology including the architecture, microarchitecture, pre-silicon validation and performance impact on media and compute-intensive applications. Also included is an intriguing paper on speculative precomputation, a technique that improves the latency of single-threaded applications by utilizing idle multithreading hardware resources to perform long-range data prefetches.

back to top
*Legal Information  |  Privacy Policy
©2002 Intel Corporation