Workshop on Resource-Efficient Cloud Computing

Half Day Workshop (morning): Sunday, June 14th, 2015 at ISCA 2015

Overview

With large-scale datacenters increasing in size and number, the challenge to improve their efficiency is becoming more pressing. Despite their prevalence, these systems remain greatly underutilized, even when techniques such as virtualization and containers are used. Given that building new datacenters or expanding existing systems is not a scalable approach to increase the compute capabilities of cloud systems moving forwards, system architects and system designers must focus on operating and managing these systems more efficiently. For this reason, over the past few years, resource-efficient cloud computing has attracted increased attention both from academia and industry. Effectively improving the resource-efficiency of large-scale systems requires cooperation among all layers of the system stack, from hardware to OSes, as well as scheduling and resource management systems and application design. The goal of the workshop is to bring together cloud computing researchers from academia and industry, underline the most pressing challenges in large-scale system design, and encourage new ways of improving the efficiency of these systems. We plan to structure the workshop as a combination of invited talks from key people in the area, and early-idea position papers.

Call for Papers

We are aiming for a mix of invited talks by experts in academia and industry, panels and peer-reviewed papers. Peer-reviewed papers will not be published in workshop proceedings, so submitting to the workshop will not preclude future publication opportunities. We especially welcome early work and position papers on future research directions. 

Topics include but are not limited to:

• Hardware support for resource partitioning and isolation
• Scheduling, resource management and cloud provisioning systems
• Software techniques for enforcing resource isolation
• Runtimes that improve utilization and application QoS
• Scalable cluster management frameworks
• Bare-metal OSes for cloud applications
• Resource-efficient application design
• Datacenter monitoring and troubleshooting
• Implications of resource-efficiency to datacenter fault-tolerance
• Real-world measurements and analysis of cloud inefficiencies

Submission Guidelines

Papers should be 4-6 pages long (8.5" x 11" pages), including figures and tables, but not including references. You may include any number of pages for references. Papers should be formatted in 2 columns, using 10-point type on 12-point leading, in a text block of 6.5" x 9". Figures and tables must be large enough to be legible when printed on 8.5" x 11" paper. Color may be used, but the paper should remain readable when printed in monochrome. Papers must be submitted for single-blind review, i.e., you can include the authors' names.
Please submit your papers over email to the workshop organizers by the submission deadline.

For any questions, please contact Christina Delimitrou at cdel@stanford.edu

Important Dates: 

Paper Submission:                      March 30 2015 23:59 PST
Author Notification:                      April 24 2015
Final Paper Submission:              May 24 2015
Workshop:                                  June 14 2015

Organizers

Christina Delimitrou, Stanford University (cdel@stanford.edu)
Prof. Christos Kozyrakis, Stanford University (kozyraki@stanford.edu)
Prof. Lingjia Tang, University of Michigan (lingjia@eecs.umich.edu)
Prof. Jason Mars, University of Michigan (profmars@eecs.umich.edu)