About cookies on this site Our websites require some cookies to function properly (required). In addition, other cookies may be used with your consent to analyze site usage, improve the user experience and for advertising. For more information, please review your options. By visiting our website, you agree to our processing of information as described in IBM’sprivacy statement. To provide a smooth navigation, your cookie preferences will be shared across the IBM web domains listed here.
Publication
SLIP 2009
Conference paper
Prediction of high-performance on-chip global interconnection
Abstract
Different interconnection structures have been proposed to solve the performance limitation caused by scaling of on-chip global wires. In this paper, we give an overview of current on-chip global interconnection structures and provide a simple model to analyze their architecture-level performance metrics. For a new category of global interconnections using on-chip transmission-line (T-line), a general framework is proposed to design and optimize such schemes. A group of experiments is performed to study and compare five different global interconnection structures in terms of latency, energy dissipation, throughput, and signal integrity across multiple technology nodes. The results show that, the T-line structures have the potential to outperform repeated RC wires at future nodes to achieve high-performance, low-power and more reliable global interconnection. Meanwhile, equalization could be helpful to improve the throughput, signal integrity, and power efficiency further. Copyright 2009 ACM.