Runtime Systems
           
Event Type Start Time End Time Rm # Chair  

 

Paper 1:30PM 2:00PM 40-41 Mary Thomas (University of Texas)
 
Title:

ParADE: An OpenMP Programming Environment for SMP Cluster Systems
  Speakers/Presenter:
Yang-Suk Kee (Seoul National University), Jin-Soo Kim (KAIST), Soonhoi Ha (Seoul National University)

 

Paper 2:00PM 2:30PM 40-41 Mary Thomas (University of Texas)
 
Title:

Dyn-MPI: Supporting MPI on Non Dedicated Clusters
  Speakers/Presenter:
D. Brent Weatherly (University of Georgia), David K. Lowenthal (University of Georgia), Mario Nakazawa (University of Georgia), Franklin Lowenthal (California State University -- Hayward)

 

Paper 2:30PM 3:00PM 40-41 Mary Thomas (University of Texas)
 
Title:

An Evaluation of a Framework for the Dynamic Load Balancing of Highly Adaptive and Irregular Parallel Applications
  Speakers/Presenter:
Kevin J. Barker (College of William and Mary), Nikos P. Chrisochoides (College of William and Mary)
             

 

     
  Session: Runtime Systems
  Title: ParADE: An OpenMP Programming Environment for SMP Cluster Systems
  Chair: Mary Thomas (University of Texas)
  Time: Wednesday, November 19, 1:30PM - 2:00PM
  Rm #: 40-41
  Speaker(s)/Author(s):  
  Yang-Suk Kee (Seoul National University), Jin-Soo Kim (KAIST), Soonhoi Ha (Seoul National University)
   
  Description:
  The demands for programming environments to exploit clusters of symmetric multiprocessors (SMPs) are increasing. In this paper, we present a new programming environment, called ParADE, to enable easy, portable, and high-performance programming on SMP clusters. It is an OpenMP programming environment on top of a multi-threaded software distributed shared memory (SDSM) system with a variant of home-based lazy release consistency (HLRC). To boost performance, the ParADE runtime system provides explicit message-passing primitives to make it a hybrid-programming environment. Collective communication primitives are used for the synchronization and work-sharing directives associated with small variables, lessening the synchronization overhead and avoiding the implicit barriers of work-sharing directives. The OpenMP translator bridges the gap between the OpenMP abstraction and the hybrid programming interfaces of the runtime system. The experiments with several NAS benchmarks and real applications on a Linux-based cluster show promising results, overcoming the poor performance of the conventional SDSM-based approaches.
  Link: Download PDF
   

 

     
  Session: Runtime Systems
  Title: Dyn-MPI: Supporting MPI on Non Dedicated Clusters
  Chair: Mary Thomas (University of Texas)
  Time: Wednesday, November 19, 2:00PM - 2:30PM
  Rm #: 40-41
  Speaker(s)/Author(s):  
  D. Brent Weatherly (University of Georgia), David K. Lowenthal (University of Georgia), Mario Nakazawa (University of Georgia), Franklin Lowenthal (California State University -- Hayward)
   
  Description:
  Distributing data is a fundamental problem in implementing efficient distributed-memory parallel programs. The problem becomes more difficult in environments where the participating nodes are not dedicated to a parallel application. We are investigating the data distribution problem in non dedicated environments in the context of explicit message-passing programs.

To address this problem, we have designed and implemented an extension to MPI called Dynamic MPI (Dyn-MPI). The key component of Dyn-MPI is its run-time system, which efficiently and automatically redistributes data on the fly when there are changes in the application or the underlying environment. Dyn-MPI supports efficient memory allocation, precise measurement of system load and computation time, and node removal. Performance results show that programs that use Dyn-MPI execute efficiently in non dedicated environments, including up to almost a three-fold improvement compared to programs that do not redistribute data and a 25% improvement over standard adaptive load balancing techniques.
  Link: Download PDF
   

 

     
  Session: Runtime Systems
  Title: An Evaluation of a Framework for the Dynamic Load Balancing of Highly Adaptive and Irregular Parallel Applications
  Chair: Mary Thomas (University of Texas)
  Time: Wednesday, November 19, 2:30PM - 3:00PM
  Rm #: 40-41
  Speaker(s)/Author(s):  
  Kevin J. Barker (College of William and Mary), Nikos P. Chrisochoides (College of William and Mary)
   
  Description:
  We present an evaluation of a flexible framework and runtime software system for the dynamic load balancing of asynchronous and highly adaptive and irregular applications. These applications, which include parallel unstructured and adaptive mesh refinement, serve as building blocks for a large class of scientific applications. Extensive study has lead to the development of solutions to the dynamic load balancing problem for loosely synchronous and computation intensive programs; however, these methods are not suitable for asynchronous and highly adaptive applications. We evaluate a new software framework which includes support for an Active Messages style communication mechanism, global name space, transparent object migration, and preemptive decision making. Our results from both a 3-dimensional parallel advancing front mesh generation program, as well as a synthetic micro-benchmark, indicate that this new framework out-performs two existing general-purpose, well-known, and widely used software systems for the dynamic load balancing of adaptive and irregular parallel applications.
  Link: Download PDF