AUTONOMIC HIGH PRODUCTIVITY COMPUTING

Overview

It is not traditional supercomputers or large clusters any more, but rather the holistic view at High Productivity Computing (HPC), which could address raw productivity or specific feature such as high performance or high availability. HPC factors in all physical (e.g., computing resources, tools, and physical space), runtime environment (e.g., workload changes, faults and automation) and human (e.g., programmers. skills and proficiency with tools) in quantifying its merit. Although the majority of performance-targeted HPC deployments are expensive custom-built solutions, they still suffer from several pain points including, but not limited to, minimal degree of autonomic capabilities, efficiency and speed of data transfer, benchmarks to evaluate their productivity and difficulty of migration to off-the-shelf commodity components.

In this research we look at establishing a High Productivity Computing (HPC) test bed environment, defining HPC metrics and targeted benchmarks and integrating autonomic middleware to enable third-party intelligence add-on. The autonomic HPC-enabling technologies that will be developed at the proposed HPC lab will enable us to efficiently: 1) optimize resources allocation in reaction to runtime workload changes for best performance and/or throughput through support for manageability, virtualization and dynamicity of runtime adaptation; 2) handle the recovery from hardware and/or software; 3) autonomize this highly dynamic environment; and 4) provide support for application workflows consisting of heterogeneous and coupled tasks/jobs through programming and runtime support for a range of computing patterns. The motivation for this project is in two folds: (i) make high-productivity the ultimate goal for high performance computing and promote its awareness; and (ii) incorporate and expose through standards autonomic middleware and tools in HPC environments to mitigate deployment and users constraints and enable automation through third party intelligence. HPC utility from the users perspective refers to the value users place on getting results through faithfully executing their tasks and guaranteeing agreed-upon Service-Level Agreements (SLAs). However from the providers perspective, HPC utility refers to dependably executing users tasks and at the same time minimizing their (providers) Total Cost of Ownership (TCO). We will Research, Develop, Test, Evaluate and integrate specific autonomics capabilities, establish metrics and benchmarks for HPC. We then measure and predict HPC efficiencies given the above constraints.

The development plan includes 1) Building an HPC Test bed to be, first, used throughout this project and later by the HPC research community as the reference test bed to experiment with and evaluate related HPC work; 2) Defining HPC metrics to measure and quantify HPC efficiency, cost and utility in an environment given a set of H/W and S/W resources, cost of running applications and the utility achieved; 3) Autonomic HPC Integration and Optimizations to support dynamic utility-driven and HPC-optimized on-demand scale-out of resources and applications, where organizations incorporate computational resources (within the enterprise and across virtual organizations) based on perceived utility; 4) Defining HPC-targeted Benchmarks to evaluate HPC techniques, algorithms and optimizations; or generally provide a quantitative measure of the productivity of a computing environment.


top 

People


Yaser Jararweh
email:
website: http://www.ece.arizona.edu/~yaser

Research Areas and Interests: Autonomic Computing & Management, Data Centers Power Management, Data Centers Virtualization Management, Data Mining,Distributed Computing, High Performance Computing, Workflow Management.


Ishtiaq Hossain
email:
website: http://ece.arizona.edu/~ihossain

Research Areas and Interests: Autonomic Computing & Management, Data Centers Power Management, Data Centers Virtualization Management, and High performance systems

top 

Publications

Salim Hariri, Yaser Jararweh, Yeliang Zhang, Talal Moukabary."Physics aware programming paradigm: approach and evaluation".CLADE’08, June 23, 2008, Boston, Massachusetts, USA.

 

top 


Sponsors

 

 

 
Phone Number: (520) 621-9915 Room 251, ECE Dept. 1230 E. Speedway Tucson, AZ 85721-0104
ACL - © Copyright 2009, Webmaster: Youssif Al-Nashif
All Rights Reserved