Navigating heterogeneous processors with market mechanisms

Published

Journal Article

Specialization of datacenter resources brings performance and energy improvements in response to the growing scale and diversity of cloud applications. Yet heterogeneous hardware adds complexity and volatility to latency-sensitive applications. A resource allocation mechanism that leverages architectural principles can overcome both of these obstacles. We integrate research in heterogeneous architectures with recent advances in multi-agent systems. Embedding architectural insight into proxies that bid on behalf of applications, a market effectively allocates hardware to applications with diverse preferences and valuations. Exploring a space of heterogeneous datacenter configurations, which mix server-class Xeon and mobile-class Atom processors, we find an optimal heterogeneous balance that improves both welfare and energy-efficiency. We further design and evaluate twelve design points along the Xeon-to-Atom spectrum, and find that a mix of three processor architectures achieves a 12× reduction in response time violations relative to equal-power homogeneous systems. © 2013 IEEE.

Full Text

Duke Authors

Cited Authors

  • Guevara, M; Lubin, B; Lee, BC

Published Date

  • July 23, 2013

Published In

Start / End Page

  • 95 - 106

International Standard Serial Number (ISSN)

  • 1530-0897

Digital Object Identifier (DOI)

  • 10.1109/HPCA.2013.6522310

Citation Source

  • Scopus