University of Virginia

The team at the University of Virginia is working on several cloud computing projects funded by the National Science Foundation (NSF). They include, Feedback-Controlled Management of Virtualized Resources for Predictable eScience and Image Super-Resolution Using Trillions of Examples.

Research Projects
Resources

Research Projects

Feedback-Controlled Management of Virtualized Resources for Predictable eScience
This project pursues a novel unified framework to ensure predictable eScience based on two dominant emerging uses of virtualized resources. The foundation of the approach is to wrap an eScience application in a performance container framework and dynamically regulate the application's performance through the application of formal feedback control theory.

Image Super-Resolution Using Trillions of Examples

Imagine continuously zooming into an image from your personal photo collection. Unlike the modern image processing software, however, this zoom operation would reveal details missing from the original image. Foe example, zooming into someone's shirt would eventually show a high-resolution image of the threads that compose it. A research team at the Department of Computer Science at the University of Virginia plans to develop techniques for intelligently enlarging a digital image that uses a database of millions of on-line images to find examples of what its components look like at a higher spatial resolution.

Resources

Paper: Self-Tuning Virtual Machines for Predictable eScience
May 19 2009
By Sang-Min Park and Marty Humphrey. Abstract: Unpredictable access to batch-mode HPC resources is a significant problem for emerging dynamic data-driven applications. Although efforts such as reservation or queue-time prediction have attempted to partially address this problem, the approaches strictly based on space-sharing impose fundamental limits on real-time predictability. In contrast, our earlier work investigated the use of feedback-controlled virtual machines (VMs), a time-sharing approach, to deliver predictable execution. However, our earlier work did not fully address usability and implementation efficiency. This paper presents an online, software-only version of feedback controlled VM, called self-tuning VM, which we argue is a practical approach for predictable HPC infrastructure. Our evaluation using five widely-used applications show our approach is both predictable and practical: by simply running time-dependent jobs with our tool, we meet a job’s deadline typically within 3% errors, and within 8% errors for the more challenging applications.

Paper: Feedback-Controlled Resource Sharing for Predictable eScience
October 22 2008
By Sang-Min Park and Marty Humphrey. Abstract: The emerging class of adaptive, real-time, data-driven applications are a significant problem for today’s HPC systems. In general, it is extremely difficult for queuing-system-controlled HPC resources to make and guarantee a tightly-bounded prediction regarding the time at which a newly-submitted application will execute. While a reservation-based approach partially addresses the problem, it can create severe resource under-utilization (unused reservations, necessary scheduled idle slots, underutilized reservations, etc.) that resource providers are eager to avoid. In contrast, this paper presents a fundamentally different approach to guarantee predictable execution. By creating a virtualized application layer called the performance container, and opportunistically multiplexing concurrent performance containers through the application of formal feedback control theory, we regulate the job’s progress such that the job meets its deadline without requiring exclusive access to resources even in the presence of a wide class of unexpected disturbances. Our evaluation using two widely-used applications, WRF and BLAST, on an 8-core server show our approach is predictable and meets deadlines with 3.4 % of errors on average while achieving high overall utilization.