Untitled

Klous, S.; Frey, Jaime; Son, Sechang; Thain, Douglas; Roy, Alain; Livny, Miron; Brand, Jo van den

doi:10.1002/cpe.v18:7

Cited by 4 publications

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

A software architecture for translucent replication

Muro

2005

Proceedings of the 2nd International Doctoral Symposium on Middleware - DSM '05

View full text Add to dashboard Cite

This paper outlines the expected research contribution of my Ph.D. thesis in the area of middleware for dependability. Dependability is a long-standing desirable property for today's network-based service applications but requirements of dependable systems are often conflicting in many ways, especially when dealing with performance related aspects. I describe a novel approach called translucent replication to address the conflicts between dependability and performance. The proposed solution is a software architecture with two key aspects: (i) enhanced intra-layer and cross-layer interaction, for flexibility and extensibility of the application/middleware infrastructure; and (ii) a semantic knowledge repository that enables smart middleware behavior, providing adaptation by matching application dependability needs, specific to data being replicated, and the system working conditions. Exploiting translucent replication, application designers can explore operating regions (instead of single points) in the dependability design-space, and dynamically investigate sets of solutions for the conflicts between dependability and performance. After presenting the general model, a realization of translucent replication within the context of a JMS solution is evaluated, to demonstrate the validity of this approach.

show abstract

A software architecture for translucent replication

Muro

2005

Proceedings of the 2nd International Doctoral Symposium on Middleware - DSM '05

View full text Add to dashboard Cite

show abstract

Environmentally responsible middleware:

Meehean

Livny

2007

Proceedings of the 16th International Symposium on High Performance Distributed Computing

View full text Add to dashboard Cite

A dynamic self-scheduling scheme for heterogeneous multiprocessor architectures

Belviranlı

Bhuyan

Gupta

2013

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Today's heterogeneous architectures bring together multiple general-purpose CPUs and multiple domainspecific GPUs and FPGAs to provide dramatic speedup for many applications. However, the challenge lies in utilizing these heterogeneous processors to optimize overall application performance by minimizing workload completion time. Operating system and application development for these systems is in their infancy. In this article, we propose a new scheduling and workload balancing scheme, HDSS, for execution of loops having dependent or independent iterations on heterogeneous multiprocessor systems. The new algorithm dynamically learns the computational power of each processor during an adaptive phase and then schedules the remainder of the workload using a weighted self-scheduling scheme during the completion phase. Different from previous studies, our scheme uniquely considers the runtime effects of block sizes on the performance for heterogeneous multiprocessors. It finds the right trade-off between large and small block sizes to maintain balanced workload while keeping the accelerator utilization at maximum. Our algorithm does not require offline training or architecture-specific parameters. We have evaluated our scheme on two different heterogeneous architectures: AMD 64-core Bulldozer system with nVidia Fermi C2050 GPU and Intel Xeon 32-core SGI Altix 4700 supercomputer with Xilinx Virtex 4 FPGAs. The experimental results show that our new scheduling algorithm can achieve performance improvements up to over 200% when compared to the closest existing load balancing scheme. Our algorithm also achieves full processor utilization with all processors completing at nearly the same time which is significantly better than alternative current approaches.

show abstract

Untitled

Cited by 4 publications

References 29 publications

A software architecture for translucent replication

A software architecture for translucent replication

Environmentally responsible middleware:

A dynamic self-scheduling scheme for heterogeneous multiprocessor architectures

Contact Info

Product

Resources

About