there axe important distributed computing systems which are neither massive nor high performance. Examples are: telecommunications systems, transportation or power networks, embedded control systems (Such as embedded ...
详细信息
ISBN:
(纸本)3540440496
there axe important distributed computing systems which are neither massive nor high performance. Examples are: telecommunications systems, transportation or power networks, embedded control systems (Such as embedded electronics in automobiles), or Systems on a Chip. Many of them are embedded system's, i.e., not directly visible to the user. For these systems, performance is not a primary issue, major issues are reviewed in this paper. then, we focus on a particular but important point, namely the correct implementation of specifications on distributed architectures.
the broad introduction of multi-core platforms into computing has brought a great opportunity to develop computationally demanding applications such as matrix computations on parallelcomputing platforms. Basic matrix...
详细信息
ISBN:
(纸本)9780769547497
the broad introduction of multi-core platforms into computing has brought a great opportunity to develop computationally demanding applications such as matrix computations on parallelcomputing platforms. Basic matrix computations such as vector and matrix addition, dot product, outer product, matrix transpose, matrix - vector and matrix multiplication are very challenging computational kernels arising in scientific computing. In this paper, we parallelize those basic matrix computations using the multi-core and parallel programming tools. Specifically, these tools are Pthreads, OpenMP, Intel Cilk++, Intel TBB, Intel ArBB, SMPSs, SWARM and FastFlow. the purpose of this paper is to present a quantitative and qualitative study of these tools for parallel matrix computations. Finally, based on this study we conclude that the Intel ArBB and SWARM parallel programming tools are the most appropriate because these give good performance and simplicity of programming.
Grid computing got much attention lately-not only from the academic world, but also from industry and business. But what remains when the dust of the many press articles has settled? We try to answer this question by ...
详细信息
ISBN:
(纸本)3540440496
Grid computing got much attention lately-not only from the academic world, but also from industry and business. But what remains when the dust of the many press articles has settled? We try to answer this question by investigating the concepts and techniques grids are based on. We distinguish three kinds of grids;the HTML-based Information Grid, the contemporary Resource Grid, and the newly evolving Service Grid. We show that grid computing is not just another hype, but has the potential to open new perspectives for the co-operative use of distributed resources. Grid computing is on the right way to solve a key problem in our distributed computing world: the discovery and coordinated use of distributed, services that may be implemented by volatile, dynamic local resources.
the model of bulk-synchronous parallel (BSP) computation is an emerging paradigm of general-purpose parallelcomputing. We propose the first optimal deterministic BSP algorithm for computingthe convex hull of a set o...
详细信息
ISBN:
(纸本)3540440496
the model of bulk-synchronous parallel (BSP) computation is an emerging paradigm of general-purpose parallelcomputing. We propose the first optimal deterministic BSP algorithm for computingthe convex hull of a set of points in three-dimensional Euclidean space. Our algorithm is based on known fundamental results from combinatorial geometry, concerning small-sized, efficiently constructible e-nets and c-approximations of a given point set. the algorithm generalises the technique of regular sampling, used previously for sorting and two-dimensional convex hull computation. the cost of the simple algorithm is optimal only for extremely large inputs;we show how to reduce the required input size by applying regular sampling in a multi-level fashion.
Heterogeneous systems, consisting of CPUs and GPUs, offer the capability to address the demands of compute- and data-intensive applications. However, programming such systems is challenging, requiring knowledge of var...
详细信息
ISBN:
(纸本)9783031506833;9783031506840
Heterogeneous systems, consisting of CPUs and GPUs, offer the capability to address the demands of compute- and data-intensive applications. However, programming such systems is challenging, requiring knowledge of various parallel programming frameworks. this paper introduces COMpar, a component-based parallel programming framework that enables the exposure and selection of multiple implementation variants of components at runtime. the framework leverages compiler directive-based language extensions to annotate the source code and generate the necessary glue code for the StarPU runtime system. COMpar provides a unified view of implementation variants and allows for intelligent selection based on runtime context. Our evaluation demonstrates the effectiveness of COMparthrough benchmark applications. the proposed approach simplifies heterogeneous parallel programming and promotes code reuse while achieving optimal performance.
Clusters that combine heterogeneous compute device architectures, coupled with novel programming models, have created a true alternative to traditional (homogeneous) cluster computing, allowing to leverage the perform...
详细信息
ISBN:
(纸本)9783642369490
Clusters that combine heterogeneous compute device architectures, coupled with novel programming models, have created a true alternative to traditional (homogeneous) cluster computing, allowing to leverage the performance of parallel applications. In this paper we introduce clOpenCL, a platform that supports the simple deployment and efficient running of OpenCL-based parallel applications that may span several cluster nodes, expanding the original single-node OpenCL model. clOpenCL is deployed through user level services, thus allowing OpenCL applications from different users to share the same cluster nodes and their compute devices. Data exchanges between distributed clOpenCL components rely on Open-MX, a high-performance communication library. We also present extensive experimental data and key conditions that must be addressed when exploiting clOpenCL with real applications.
parallelism permeates all levels of current computing systems, from single CPU machines, to large server farms, to geographically dispersed “volunteers-who collaborate over the Internet. the effective use of parallel...
详细信息
ISBN:
(纸本)9783642328206
parallelism permeates all levels of current computing systems, from single CPU machines, to large server farms, to geographically dispersed “volunteers-who collaborate over the Internet. the effective use of parallelism depends crucially on the availability of faithful, yet tractable, models of computation for algorithm design and analysis, and on efficient strategies for solving key computational problems on prominent classes of computing platforms. No less important are good models of the way the different components/subsystems of a platform are interconnected. Withthe development of new genres of computing platforms, such as multicore parallel machines, desktop grids, clouds, and hybrid GPU/CPUbased systems, new models and paradigms are needed that will allow parallel programming to advance into mainstream computing. Topic 12 focuses on contributions providing new results on foundational issues regarding parallelism in computing, and/or proposing improved approaches to the solution of specific algorithmic problems.
Since 2004 the research interests of the CoreGRID community has evolved from distributed large scale computing to service-based computing and Clouds. the adoption of the SOA paradigm and virtualization has resulted in...
详细信息
ISBN:
(纸本)9783642218781;9783642218774
Since 2004 the research interests of the CoreGRID community has evolved from distributed large scale computing to service-based computing and Clouds. the adoption of the SOA paradigm and virtualization has resulted in an unprecedented flexibility in creating distributed applications. Old and new research challenges need to be mastered to exploit fully the potential of cloud infrastructures. In this article we present outstanding cloud-related research questions that need to be addressed. It is proposed that a pan-european research community is needed to bridge existing knowledge gaps.
It becomes obvious that traditional platforms and processing paradigms can't store and process huge amounts of data. the only solution is to use specially designed ad-hoc platform/architecture based on paralleliza...
详细信息
ISBN:
(纸本)9781728173863
It becomes obvious that traditional platforms and processing paradigms can't store and process huge amounts of data. the only solution is to use specially designed ad-hoc platform/architecture based on parallelization that distributes data across large cluster of physical machines. Data Intensive computing is a subclass of general parallelcomputing concept which is based on division of large amounts of data into independent parts and processing them in parallel. In the paper the alternative parallelization architectures are reviewed. MapReduce Programming model associated with distributed massive parallel processing of large amount of data is examined. the main objective of this study is to investigate conceptual fundament behind very popular data-drive computation model MapReduce.
Concurrent hash tables are one of the fundamental building blocks for cloud computing. In this paper, we introduce lock-free modifications to in-memory bucketized cuckoo hashing. We present a novel concurrent strategy...
详细信息
ISBN:
(纸本)9783031396977;9783031396984
Concurrent hash tables are one of the fundamental building blocks for cloud computing. In this paper, we introduce lock-free modifications to in-memory bucketized cuckoo hashing. We present a novel concurrent strategy in designing a lock-free hash table, called LFBCH, that paves the way towards scalability and high space efficiency. To the best of our knowledge, this is the first attempt to incorporate lock-free technology into in-memory bucketized cuckoo hashing, while still providing worst-case constant-scale lookup time and extremely high load factor. All of the operations over LFBCH, such as get, put, "kick out" and rehash, are guaranteed to be lock-free, without introducing notorious problems like false miss and duplicated key. the experimental results indicate that under mixed workloads with up to 64 threads, the throughput of LFBCH is 14%-360% higher than other popular concurrent hash tables.
暂无评论