In this paper, we introduce the Join-Elect-Leave (JEL) model, a simple yet powerful model for tracking the resources participating in an application. This model is based on the concept of signaling, i.e., notifying th...
详细信息
ISBN:
(纸本)9781595939975
In this paper, we introduce the Join-Elect-Leave (JEL) model, a simple yet powerful model for tracking the resources participating in an application. This model is based on the concept of signaling, i.e., notifying the application when resources have Joined or Left the computation. In addition, the model includes Elections, which can be used to select resources with a special role. JEL supports several consistency models and is suitable for resource coordination of a wide variety of applications, ranging from the traditional fixed resource sets used in MPI, to flexible grid-oriented programming models.
In this paper we consider productivity challenges for parallel programmers and explore ways that parallel language design might help improve end-user productivity. We offer a candidate list of desirable qualities for ...
详细信息
In this paper we consider productivity challenges for parallel programmers and explore ways that parallel language design might help improve end-user productivity. We offer a candidate list of desirable qualities for a parallel programming language, and describe how these qualities are addressed in the design of the Chapel language. In doing so, we provide an overview of Chapel's features an how they help address parallel productivity. We also survey current techniques for parallel programming and describe ways in which we consider them to fall short of our idealized productive programming model.
Efficient programming of multimedia streaming applications for Consumer Electronics (CE) devices is not trivial. As a solution for this problem, we present SP@CE, a novel programming model designed to balance the spec...
详细信息
ISBN:
(纸本)9783540725206
Efficient programming of multimedia streaming applications for Consumer Electronics (CE) devices is not trivial. As a solution for this problem, we present SP@CE, a novel programming model designed to balance the specific requirements of CE streaming applications with the simplicity and efficiency of the Series-Parallel Contention (SPC) programming model. To enable the use of SP@CE, we have designed a framework that guides the programmer to design, evaluate, optimize and execute the application on the target CE platform. To evaluate the entire system, we have used SP@CE to implement a set of real-life streaming applications and we present the results obtained by running them on the Wasabi/SpaceCAKE architecture from Philips, a multiprocessor system-on-chip (MPSoC) CE platform. The experiments show that SP@CE enables rapid application development, induces low overhead, offers high code reuse potential, and takes advantage of the inherent application parallelism.
This paper addresses the problem of orchestrating and scheduling parallelism at multiple levels of granularity on heterogeneous multicore processors. We present mechanisms and policies for adaptive exploitation and sc...
详细信息
ISBN:
(纸本)9781595936028
This paper addresses the problem of orchestrating and scheduling parallelism at multiple levels of granularity on heterogeneous multicore processors. We present mechanisms and policies for adaptive exploitation and scheduling of layered parallelism on the Cell Broadband Engine. Our policies combine event-driven task scheduling with malleable loop-level parallelism, which is exploited from the runtime system whenever task-level parallelism leaves idle cores. We present a scheduler for applications with layered parallelism on Cell and investigate its performance with RAxML, an application which infers large phylogenetic trees, using the Maximum Likelihood (ML) method. Our experiments show that the Cell benefits significantly from dynamic methods that selectively exploit the layers of parallelism in the system, in response to workload fluctuation. Our scheduler outperforms the MPI version of RAxML, scheduled by the Linux kernel, by up to a factor of 2.6. We are able to execute RAxML on one Cell four times faster than on a dual-processor system with Hyperthreaded Xeon processors, and 5-10% faster than on a single-processor system with a dual-core, quad-thread IBM Power5 processor.
A policy maker knows two models. One implies an exploitable inflation-unemployment trade-off, the other does not. The policy maker's prior probability over the two models is part of his state vector. Bayes' la...
详细信息
A policy maker knows two models. One implies an exploitable inflation-unemployment trade-off, the other does not. The policy maker's prior probability over the two models is part of his state vector. Bayes' law converts the prior probability into a posterior probability and gives the policy maker an incentive to experiment. For models calibrated to U.S. data through the early 1960s, we compare the outcomes from two Bellman equations. The first tells the policy maker to "experiment and learn." The second tells him to "learn but don't experiment." In this way, we isolate a component of government policy that is due to experimentation and estimate the benefits from intentional experimentation. We interpret the Bellman equation that learns but does not intentionally experiment as an "anticipated utility" model and study how well its outcomes approximate those from the "experiment and learn" Bellman equation. The approximation is good. For our calibrations, the benefits from purposeful experimentation are small because random shocks are big enough to provide ample unintentional experimentation.
This paper describes the Aggregate Remote Memory Copy Interface (ARMCI), a portable high performance remote memory access communication interface, developed originally under the U.S. Department of Energy (DOE) Advance...
详细信息
This paper describes the Aggregate Remote Memory Copy Interface (ARMCI), a portable high performance remote memory access communication interface, developed originally under the U.S. Department of Energy (DOE) Advanced Computational Testing and Simulation Toolkit project and currently used and advanced as a part of the run-time layer of the DOE project, programming models for Scalable Parallel Computing. The paper discusses the model, addresses challenges of portable implementations, and demonstrates that ARMCI delivers high performance on a variety of platforms. Special emphasis is placed on the latency hiding mechanisms and ability to optimize noncontiguous data transfers.
Botswana has vast proven deposits of steam coal, which, for a long time, the government has wanted to develop but without much success. The main objectives of this study are to forecast possible coal exports from Bots...
详细信息
Botswana has vast proven deposits of steam coal, which, for a long time, the government has wanted to develop but without much success. The main objectives of this study are to forecast possible coal exports from Botswana and the land routes for these exports to determine the competitiveness of Botswana's coal in world steam coal trade to make recommendations on the appropriate policy for the exploitation of this coal. To accomplish these objectives, we construct a model of the global steam coal trade and apply this model to forecast the likely optimal size of mine, timing of capacity, and choice of export port for the years 2005 and 2010 from a 2000 base forecast year. The results of our regional analysis suggest that Botswana's coal exports are competitive in Asia and Western Europe. These results are shown to be least sensitive to changes in rail transportation costs and marginal supply costs but more sensitive to changes in capital costs for mine development.
Sensor networks open a new frontier for embedded-distributed computing. Paradigms for sensor network programming-in-the-large have been identified as a significant challenge toward developing large-scale applications....
详细信息
The rise of SOA and PGAS doesn't quite spell the end of Corba, RMI, and other classic distributed-object systems, but it does increasingly restrict their ranges of use. These technologies surely still play major r...
详细信息
The rise of SOA and PGAS doesn't quite spell the end of Corba, RMI, and other classic distributed-object systems, but it does increasingly restrict their ranges of use. These technologies surely still play major roles in enterprise-integration systems and heterogeneous intranet applications and niches such as real-time Corba, as well as the countless existing deployments that remain with us for years. They might even reemerge as the approach of choice in future grid computing frameworks that must sometimes straddle administrative domains. But the idea that one approach to object-oriented middleware could serve all needs seems to be one whose time has come and gone
暂无评论