Modern GPUs offer an accessible and scalable supercomputing platform. However, their use for state-space search has so far been auxiliary or in niche methods such as iterative deepening A* or random walks. We describe...
详细信息
ISBN:
(纸本)9781728189468
Modern GPUs offer an accessible and scalable supercomputing platform. However, their use for state-space search has so far been auxiliary or in niche methods such as iterative deepening A* or random walks. We describe a scalable deterministic A* state space search implementation running entirely on the GPU. This is made possible by designing core algorithms for lock-step parallelism as well as using CUDA cooperative grids to implement hash-distributed A* load balancing scheme. Comparative benchmarking showed significant improvements in performance as well as resource efficiency.
As distributed applications became more commonplace and more sophisticated, new programming languages and models for distributed programming were created. In the context of continuously increasing data flows in parall...
详细信息
ISBN:
(纸本)9781479959198
As distributed applications became more commonplace and more sophisticated, new programming languages and models for distributed programming were created. In the context of continuously increasing data flows in parallel applications, there is a renewed interest in the dataflow paradigm. This paper shows why AGAPIA language is suitable for dataflow programming. AGAPIA is capable of expressing massive parallelism in a manageable way for programmers, allowing building dynamic nodes and links in the data flow graph at runtime. The nodes of the dataflow graph, also called programs, are modular and reusable. The communication is transparent for users allowing them to concentrate on the high level flow and algorithm. A complete application and an analysis in terms of productivity and performance are presented in order to demonstrate the AGAPIA's capabilities.
Proper use of land require tools to handle each parcel both in a individual manner, as regard to the activities rolled out on it, and as part of a farm or greenhouse containing several parcels, in order to plan and ha...
详细信息
ISBN:
(纸本)9781467371483
Proper use of land require tools to handle each parcel both in a individual manner, as regard to the activities rolled out on it, and as part of a farm or greenhouse containing several parcels, in order to plan and harness the results of all activities in a farm or greenhouse. The paper presents services related to soil management, within a platform that provides integrated support for greenhouse activities, which are subject of CLUeFARM research project (Information system based on cloud services, accessible through mobile devices, for quality improvement of products and business development in farms).
How can the CS education community shift to teach more parallel and distributedcomputing (PDC) soon enough to prepare its graduates for their careers? The CSinparallel project implements strategies that would accompl...
详细信息
ISBN:
(纸本)9780769546766
How can the CS education community shift to teach more parallel and distributedcomputing (PDC) soon enough to prepare its graduates for their careers? The CSinparallel project implements strategies that would accomplish such a shift. An Early Adopter project to correlate CSinparallel materials and NSF/TCPP Curricular Recommendations is described. Citing research on scalability of innovations, such synergistic collaborations offer hope for scalably and rapidly adding PDC to undergraduate CS education.
Public resource computing depends on the availability of computing resources that have been contributed by individuals. The amount of resources can be increased by incentivizing resource providers through payment for ...
详细信息
ISBN:
(纸本)9781424437511
Public resource computing depends on the availability of computing resources that have been contributed by individuals. The amount of resources can be increased by incentivizing resource providers through payment for resources. However, there are costs associated with providing resources for public grid computing. Without an understanding of these costs, it is impossible for a provider to judge if the payment is sufficient to overcome those costs. In this paper, we present a provider cost model that considers all resource provider costs including opportunity costs, future-value costs, penalties, utility costs, and fixed costs. This model helps set a cost structure that a resource provider can use to determine whether it is profitable to participate in a public resource computing market.
The ubiquity of multicore processors, cloud computing, and hardware accelerators have elevated parallel and distributedcomputing (PDC) topics into fundamental building blocks of the undergraduate CS curriculum. There...
详细信息
ISBN:
(纸本)9781665497473
The ubiquity of multicore processors, cloud computing, and hardware accelerators have elevated parallel and distributedcomputing (PDC) topics into fundamental building blocks of the undergraduate CS curriculum. Therefore, it is increasingly important for students to learn a common core of introductory PDC topics and develop parallel thinking skills early in their CS studies. We present the curricular design, pedagogy, and goals of an introductory-level course on computer systems that introduces parallelcomputing to students who have only a CS1 background. Our course focuses on three curricular goals that serve to integrate the ACM-IEEE TCPP guidelines throughout: a vertical slice through the computer of how it runs a program;evaluating system costs associated with running a program;and taking advantage of the power of parallelcomputing. We elaborate on the goals and details of our course's key modules, and we discuss our pedagogical approach that includes active-learning techniques. We find that the PDC foundation gained through early exposure in this course helps students gain confidence in their ability to expand and apply their understanding of PDC concepts throughout their CS education.
The main goal of this workshop is to provide a timely forum for the exchange and dissemination of new ideas, techniques and research in the field of the parallel and distributed computational models. The workshop is m...
The main goal of this workshop is to provide a timely forum for the exchange and dissemination of new ideas, techniques and research in the field of the parallel and distributed computational models. The workshop is meant to bring together researchers and practitioners interested in all aspects of parallel and distributedcomputing taken in an inclusive, rather than exclusive, sense. We are convinced that the workshop atmosphere will be conducive to open and mutually beneficial exchanges of ideas between the participants.
Reasoning about, and debugging, hierarchical control plane systems is hard. Moreover, OpenTracing, the industry adopted tracing model, has problems with tracing activities in presence of coalescing effects, which mate...
详细信息
ISBN:
(纸本)9781665432818
Reasoning about, and debugging, hierarchical control plane systems is hard. Moreover, OpenTracing, the industry adopted tracing model, has problems with tracing activities in presence of coalescing effects, which materialize, among others, in cloud platforms and build systems. In our earlier work we have proposed a novel approach to distributed systems tracing, based on an extension of OpenTracing. The aim of this contribution is to outline how the proposed approach can be implemented.
Massive parallel computational fluid dynamics is a vital tool for the modern aviation industry. The parallel aerodynamics numerical simulation software based on unstructured mixed grid and cell-centered finite volume ...
详细信息
ISBN:
(纸本)9781509041527
Massive parallel computational fluid dynamics is a vital tool for the modern aviation industry. The parallel aerodynamics numerical simulation software based on unstructured mixed grid and cell-centered finite volume method (FVM) was presented in this paper, which was developed for the complicated configurations of industrial grade. The algorithms for parallel implementation with message passing interface (MPI), including compact numerical discrete schemes, grid partition method for multiprocessors and data transfer scheme, are discussed in detail. The validity of parallelcomputing is verified using a revolving model of relatively small grid. Then a large transport airplane configuration with hundred millions of grid elements is used for parallel testing, and the efficiency is preserved above 80% up to 18816 processors.
Building correct distributed systems is challenging, and any attempt for providing a direct, global proof of correctness of a distributed system is bound to fail. An interesting alternative approach consists in starti...
详细信息
ISBN:
(纸本)9781479959198
Building correct distributed systems is challenging, and any attempt for providing a direct, global proof of correctness of a distributed system is bound to fail. An interesting alternative approach consists in starting from a specification or program of the system under construction, verifying all properties of interest on it - which has a much lower complexity than the verification on a distributed implementation - and finally derive a distributed implementation using some correct by-construction approach. Note that this topic is related to distributed control, where the objective is to enforce in a distributed manner some global constraint on a plant. Deriving such a distributed controller directly is difficult, and the correctness of the resulting controller is difficult to prove. A more feasible approach in this context is to first construct a global controller, then transform it into distributed one, again by means of a correct-by-construction approach.
暂无评论