检索结果-内蒙古大学图书馆

Development of SCIT Supercomputers Family Created at the V. M. Glushkov Institute of Cybernetics, NAS of Ukraine, in 2002-2017

学校读者我要写书评

暂无评论

CYBERNETICS AND SYSTEMS ANALYSIS 2017年第4期53卷 600-604页

作者： Golovynskyi, A. Sergienko, I. Tulchinskyi, V. Malenko, A. Bandura, O. Gorenko, S. Roganova, O. Lavrikova, O. Natl Acad Sci Ukraine M Glushkov Inst Cybernet Kiev Ukraine Natl Acad Sci VM Glushkov Inst Cybernet Kuiv Ukraine

The year 2017 marks the 15th anniversary of the SCIT supercomputer project, which allows us to summarize the results and draw conclusions. In this paper, we discuss the evolution of SCIT architecture and statistics of the supercomputing center for years 2002-2017. These data will be useful for computer cluster developers and researchers who design resource management algorithms for computing clusters.

关键词： supercomputer cluster parallel programming

Resilient blocks for summarising distributed data

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Audrito, Giorgio Bergamini, Sergio University of Torino Italy

Summarising distributed data is a central routine for parallel programming, lying at the core of widely used frameworks such as the map/reduce paradigm. In the IoT context it is even more crucial, being a privileged mean to allow long-range interactions: in fact, summarising is needed to avoid data explosion in each computational unit. We introduce a new algorithm for dynamic summarising of distributed data, weighted multi-path, improving over the state-of-the-art multi-path algorithm. We validate the new algorithm in an archetypal scenario, taking into account sources of volatility of many sorts and comparing it to other existing implementations. We thus show that weighted multi-path retains adequate accuracy even in high-variability scenarios where the other algorithms are diverging significantly from the correct values. Copyright © 2018, The Authors. All rights reserved.

关键词： parallel programming

学校读者我要写书评

暂无评论

HPCS 2018 Tutorials

HPCS 2018 Tutorials

International Conference on High Performance Computing & Simulation (HPCS)

Provides an abstract of the tutorial presentation and may include a brief professional biography of the presenter. The complete presentation was not made available for publication as part of the conference proceedings.

关键词： Tutorials Computational modeling Containers Runtime Task analysis parallel programming

ROME 2018 Invited Talk

学校读者我要写书评

暂无评论

ROME 2018 Invited Talk

IEEE International Symposium on parallel and Distributed Processing Workshops and Phd Forum (IPDPSW)

作者： Karl Fuerlinger

Provides an abstract of the invited presentation and may include a brief professional biography of the presenter. The complete presentation was not made available for publication as part of the conference proceedings.

ISBN: (纸本)9781538655566;9781538655559

关键词： Electronics packaging Runtime Distributed processing Conferences Tools parallel programming

A NUMA-aware provably-efficient task-parallel platform based on the work-first principle

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Deters, Justin Wu, Jiaye Xu, Yifan Lee, I-Ting Angelina Washington University St. Louis United States

Task parallelism is designed to simplify the task of parallel programming. When executing a task parallel program on modern NUMA architectures, it can fail to scale due to the phenomenon called work inflation, where the overall processing time that multiple cores spend on doing useful work is higher compared to the time required to do the same amount of work on one core, due to effects experienced only during parallel executions such as additional cache misses, remote memory accesses, and memory bandwidth issues. One can mitigate work inflation by co-locating the computation with its data, but this is nontrivial to do with task parallel programs. First, by design, the scheduling for task parallel programs is automated, giving the user little control over where the computation is performed. Second, the platforms tend to employ work stealing, which provides strong theoretical guarantees, but its randomized protocol for load balancing does not discern between work items that are far away versus ones that are closer. In this work, we propose NUMA-WS, a NUMA-aware task parallel platform engineered based on the work-first principle. By abiding by the work-first principle, we are able to obtain a platform that is work efficient, provides the same theoretical guarantees as a classic work stealing scheduler, and mitigates work inflation. We have extended Cilk Plus runtime system to implemented NUMA-WS. Empirical results indicate that the NUMA-WS is work efficient and can provide better scalability by mitigating work inflation. Copyright © 2018, The Authors. All rights reserved.

关键词： parallel programming

A minimalistic approach for fast computation of geodesic distances on triangular meshes

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Calla, Luciano A. Romero Perez, Lizeth J. Fuentes Montenegro, Anselmo A. University of Zurich Switzerland Institute of Computing Federal Fluminense University Niteroi Brazil IPRODAM3D Research Group La Salle University Arequipa Peru

The computation of geodesic distances is an important research topic in Geometry Processing and 3D Shape Analysis as it is a basic component of many methods used in these areas. In this work, we present a minimalistic parallel algorithm based on front propagation to compute approximate geodesic distances on meshes. Our method is practical and simple to implement, and does not require any heavy pre-processing. The convergence of our algorithm depends on the number of discrete level sets around the source points from which distance information propagates. To appropriately implement our method on GPUS taking into account memory coalescence problems, we take advantage of a graph representation based on a breadth-first search traversal that works harmoniously with our parallel front propagation approach. We report experiments that show how our method scales with the size of the problem. We compare the mean error and processing time obtained by our method with such measures computed using other methods. Our method produces results in competitive times with almost the same accuracy, especially for large meshes. We also demonstrate its use for solving two classical geometry processing problems: The regular sampling problem and the Voronoi tessellation on meshes. Copyright © 2018, The Authors. All rights reserved.

关键词： parallel programming

parallel Algorithms for Bitmap Image Processing Based on Daubechies Wavelets

学校读者我要写书评

暂无评论

Parallel Algorithms for Bitmap Image Processing Based on Dau...

International Conference on Communication Software and Networks, ICCSN

作者： Utkir Khamdamov Hakimjon Zaynidinov Dept. Hardware and Software of control systems in Telecommunication Tashkent University of Information Technologies Tashkent Uzbekistan Dept. Information Technologies Tashkent University of Information Technologies Tashkent Uzbekistan

This paper is devoted to the research of bitmap image processing based on wavelet functions. The Daubechies wavelet function was used as a mathematical model for filtering, compression and smoothing of two-dimensional signals, because the analysis of existing wavelet functions showed that the Daubechies wavelet family is most effective for image processing. OpenMP parallel programming in C/C++ was used for the parallelization of computing processes in image processing problems.

关键词： Message systems Wavelet transforms Image processing parallel programming Instruction sets parallel processing Computational modeling

Programmable switch as a parallel computing device

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Chen, Li Chen, Ge Lingys, Justinas Chen, Kai SING Group CSE HKUST

Modern switches have packet processing capacity of up to multi-tera bits per second, and they are also becoming more and more programmable. We seek to understand whether the programmability can translate packet processing capacity to computational power for parallel computing applications. In this paper, we first develop a simple mathematical model to understand the costs and overheads of data plane computation. Then we validate the the performance benefits of offloading computation to network. Using experiments on real data center network, we find that offloading computation to the data plane results in up to 20× speed-up for a simple Map-Reduce application. Motivated by this, we propose a parallel programming framework, p4mr, to help users efficiently program multiple switches. We successfully build and test a prototype of p4mr on a simulated testbed. Copyright © 2018, The Authors. All rights reserved.

关键词： parallel programming

Toward a versatile and scalable metadata distribution framework for object storage

学校读者我要写书评

暂无评论

Toward a versatile and scalable metadata distribution framew...

International Conference on High Performance Computing and Simulation

作者： Eloise BILLA Soraya ZERTAL Thomas LEIBOVICI Philippe DENIEL DAM UVSQ Versailles France Li-PaRAD Univ. of Versailles Versailles France DAM CEA Arpajon France

ISBN: (纸本)9781538678800

The use of object storage in the HPC world becomes a common case as it enables to overcome some POSIX limitations in scalability and performance. Indeed, object stores use a flat namespace, avoiding hierarchy in access requests and the cost of maintaining dependencies between multiple entries. Object stores also differentiate data flow from metadata flow, providing better concurrency and throughput. They can store trillions of objects and each object has its own customized metadata attributes so these metadata can be richer than POSIX attributes.

关键词： Metadata Servers Software Scalability Throughput Concurrent computing Synchronization Metadata Scalability Throughput Servers computer software parallel programming Unix operating system shopping mall data flow