Software portability

About: Software portability is a research topic. Over the lifetime, 8987 publications have been published within this topic receiving 164922 citations. The topic is also known as: portability.

Smart multi-task scheduling for OpenCL programs on CPU/GPU heterogeneous platforms

Abstract: Heterogeneous systems consisting of multiple CPUs and GPUs are increasingly attractive as platforms for high performance computing. Such platforms are usually programmed using OpenCL which provides program portability by allowing the same program to execute on different types of device. As such systems become more mainstream, they will move from application dedicated devices to platforms that need to support multiple concurrent user applications. Here there is a need to determine when and where to map different applications so as to best utilize the available heterogeneous hardware resources. In this paper, we present an efficient OpenCL task scheduling scheme which schedules multiple kernels from multiple programs on CPU/GPU heterogeneous platforms. It does this by determining at runtime which kernels are likely to best utilize a device. We show that speedup is a good scheduling priority function and develop a novel model that predicts a kernel's speedup based on its static code structure. Our scheduler uses this prediction and runtime input data size to prioritize and schedule tasks. This technique is applied to a large set of concurrent OpenCL kernels. We evaluated our approach for system throughput and average turn-around time against competitive techniques on two different platforms: a Core i7/Nvidia GTX590 and a Core i7/AMD Tahiti 7970 platforms. For system throughput, we achieve, on average, a 1.21x and 1.25x improvement over the best competitors on the NVIDIA and AMD platforms respectively. Our approach reduces the turnaround time, on average, by at least 1.5x and 1.2x on the NVIDIA and AMD platforms respectively, when compared to alternative approaches.

Evaluation of Architectures for Mobile Robotics

Abstract: In this paper we make a comparative study of some successful software architectures for mobile robot systems. The objective is to gather experience for the future design of a new robot architecture. Three architectures are studied more closely, Saphira, TeamBots and BERRA. Qualities such as portability, ease of use, software characteristics, programming and run-time efficiency are evaluated. In order to get a true hands-on evaluation, all the architectures are implemented on a common hardware robot platform. A simple reference application is made with each of these systems. All the steps necessary to achieve this are discussed and compared. Run-time data are also gathered. Conclusions regarding the results are made, and a sketch for a new architecture is made based on these results.

Scalable and massively parallel Monte Carlo photon transport simulations for heterogeneous computing platforms

Abstract: We present a highly scalable Monte Carlo (MC) three-dimensional photon transport simulation platform designed for heterogeneous computing systems. Through the development of a massively parallel MC algorithm using the Open Computing Language framework, this research extends our existing graphics processing unit (GPU)-accelerated MC technique to a highly scalable vendor-independent heterogeneous computing environment, achieving significantly improved performance and software portability. A number of parallel computing techniques are investigated to achieve portable performance over a wide range of computing hardware. Furthermore, multiple thread-level and device-level load-balancing strategies are developed to obtain efficient simulations using multiple central processing units and GPUs.

Robust Registration of Multi-modal Images: Towards Real-Time Clinical Applications

Abstract: High performance computing has become a key step to introduce computer tools, like real-time registration, in the medical field. To achieve real-time processing, one usually simplifies and adapts algorithms so that they become application and data specific. This involves designing and programming work for each application, and reduces the generality and robustness of the method. Our goal in this paper is to show that a general registration algorithm can be parallelized on an inexpensive and standard parallel architecture with a mall amount of additional programming work, thus keeping intact the algorithm performance.For medical applications, we show that a cheap cluster of dual-processor PCs connected by an Ethernet network is a good trade-off between the power and the cost of the parallel platform. Portability, scalability and safety requirements led us to choose OpenMP to program multiprocessor machines and MPI to coordinate the different nodes of the cluster. The resulting computation times are very good on small and medium resolution images, and they are still acceptable on high resolution MR images (resp. 19, 45 and 95 seconds on 5 dual-processors Pentium III 933 MHz).

Chasing Stars: The Myth of Talent and the Portability of Performance

Abstract: Acknowledgments ix Introduction 3 Part One: Talent and Portability Chapter 1: Moving On 15 Chapter 2: Analysts' Labor Market 35 Chapter 3: The Limits of Portability 51 Chapter 4: Do Firms Benefit from Hiring Stars? 77 Part Two: Facets of Portability Chapter 5: Stars and Their Galaxies: Firms of Origin and Portability 93 Chapter 6: Integrating Stars: The Hiring Firm and Portability of Performance 125 Chapter 7: Liftouts (Taking Some of It with You): Moving in Teams 141 Chapter 8: Women and Portability: Why Is Women's Performance More Portable than Men's? 163 Part Three: Implications for Talent Management: Developing, Retaining, and Rewarding Stars Chapter 9: Star Formation: Developmental Cultures at Work 197 Chapter 10: Turnover: Who Leaves and Why 239 Chapter 11: A Special Case of Turnover: Stars as Entrepreneurs 253 Chapter 12: Measuring and Rewarding Stars' Performance 273 Chapter 13: Lessons from Wall Street and Elsewhere 321 Appendix 343 Notes 353 Index 437