Continuous circuit miniaturization and increased process variability point to a future with diminishing returns from dynamic voltage scaling. Operation below Vcc-min has been proposed recently as a mean to reverse this trend. The goal of this paper is to minimize the performance loss due to reduced cache capacity when operating below Vcc-min. A simple method is proposed: disable faulty blocks at low voltage. The method is based on observations regarding the distributions of faults in an array according to probability theory. The key lesson, from the probability analysis, is that as the number of uniformly distributed random faulty cells in an array increases the faults increasingly occur in already faulty blocks. The probability analysis is also shown to be useful for obtaining insight about the reliability implications of other cache techniques.For one configuration used in this paper, block disabling is shown to have on the average 6.6% and up to 29% better performance than a previously proposed scheme for low voltage cache operation. Furthermore, block-disabling is simple and less costly to implement and does not degrade performance at or above Vcc-min operation. Finally, it is shown that a victim-cache enables higher and more deterministic performance for a block-disabled cache.
While reproducing the experimental results of research articles is standard practice in mature domains of science, such as physics or biology, it has not yet become mainstream in computer architecture. However, recent research shows that the lack of a fair and broad comparison of research ideas can be significantly detrimental to the progress, and thus the productivity, of research. At the same time, the complexity of architecture simulators and the fact that simulators are not systematically disseminated with novel ideas are largely responsible for this situation. While this methodology has a fundamental impact on research, it is by essence a practical issue.In this article, we present and set up an atypical approach to overcome this practical methodology issue, which takes the form of an open and continuous exploration through ArchExplorer, a server-side web infrastructure, that can significantly ease the process of fairly and quantitatively comparing research ideas. The web infrastructure ArchExplorer.org is now publicly open, and we demonstrate the approach with a set of data cache mechanisms. We show that this broad exploration can challenge some earlier assessments about data cache research, and even challenges the conclusions of an earlier but less thorough study on data cache comparison.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.