检索结果-内蒙古大学图书馆

25th International Middleware Conference

作者： Tan, Yuesheng Yang, Lei Li, Wenhao Wu, Yuda South China Univ Technol Guangzhou Guangdong Peoples R China

ISBN: (纸本)9798400706233

Edge AI aims to enable distributed machine learning (DML) on edge resources to fulfill the need for data privacy and low latency. Meanwhile, the challenge of device heterogeneity and discrepancy in data distribution requires more sophisticated DML architectures that differ in topology and communication strategy. This calls for a standardized and general programming interface and framework to develop them. Existing frameworks are only meant for specific architectures (e.g., FedML and Flower for Federated Learning) and do not support others by design. This paper presents RoleML, a novel programming model for general DML architecture development. RoleML breaks a DML architecture into a series of interactive components and represents them with a unified abstraction named role. The behavior of each role is expressed as several message channels, while its workloads are declared as elements in order to decouple from the distributed training workflow. Powered by a runtime system, RoleML allows developers to flexibly and dynamically assign roles to different computation nodes, simplifying the implementation of complex architectures. The runtime further improves the reliability of DML applications via an automatic role offloading mechanism based on containerization. We conduct case studies and experiments to demonstrate the wide applicability and high customizability of RoleML with low performance overhead, as well as the effectiveness of the role offloading mechanism in mitigating device overload.

关键词： programming model Distributed Machine Learning Edge Computing Computation Offloading

来源：评论

学校读者我要写书评

暂无评论

Lightning: Scaling the GPU programming model Beyond a Single GPU 36

Lightning: Scaling the GPU Programming Model Beyond a Single...

引用

36th IEEE International Parallel and Distributed Processing Symposium (IEEE IPDPS)

作者： Heldens, Stijn Hijma, Pieter Werkhoven, Ben Van Maassen, Jason van Nieuwpoort, Rob, V Netherlands eSci Ctr Amsterdam Netherlands Univ Amsterdam Amsterdam Netherlands Vrije Univ Amsterdam Amsterdam Netherlands

ISBN: (纸本)9781665481069

The GPU programming model is primarily aimed at the development of applications that run one GPU. However, this limits the scalability of GPU code to the capabilities of a single GPU in terms of compute power and memory capacity. To scale GPU applications further, a great engineering effort is typically required: work and data must be divided over multiple GPUs by hand, possibly in multiple nodes, and data must be manually spilled from GPU memory to higher-level memories. We present Lightning: a framework that follows the common GPU programming paradigm but enables scaling to large problems with ease. Lightning supports multi-GPU execution of GPU kernels, even across multiple nodes, and seamlessly spills data to higher-level memories (main memory and disk). Existing CUDA kernels can easily be adapted for use in Lightning, with data access annotations on these kernels allowing Lightning to infer their data requirements and the dependencies between subsequent kernel launches. Lightning efficiently distributes the work/data across GPUs and maximizes efficiency by overlapping scheduling, data movement, and kernel execution when possible. We present the design and implementation of Lightning, as well as experimental results on up to 32 GPUs for eight benchmarks and one real-world application. Evaluation shows excellent performance and scalability, such as a speedup of 57.2x over the CPU using Lighting with 16 GPUs over 4 nodes and 80 GB of data, far beyond the memory capacity of one GPU.

关键词： GPU distributed computing CUDA programming model

来源：评论

学校读者我要写书评

暂无评论

Convergence of HPC and Big Data in extreme-scale data analysis through the DCEx programming model 34

Convergence of HPC and Big Data in extreme-scale data analys...

引用

34th IEEE International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD)

作者： Garcia-Blas, Javier Fernandez Munoz, Javier Carretero, Jesus Marozzo, Fabrizio Talia, Domenico Trunfio, Paolo Fernandez-Pena, Alberto Martin de Blas, Daniel Univ Carlos III Madrid Comp Sci & Engn Dept Leganes Spain Univ Calabria DIMES Dept Arcavacata Di Rende Italy Univ Carlos III Madrid Inst Invest Sanitaria Gregorio Maranon Dept Bioingn & Ingn Aerosp Madrid Spain

ISBN: (数字)9781665451550

ISBN: (纸本)9781665451550

High-level programming models can help application developers to access and use resources without the need to manage low-level architectural entities, as a parallel programming model defines a set of programming abstractions that simplify the way by which a programmer structures and expresses her/his algorithm. Early proposals of Exascale programming tools are based on the adaptation of traditional parallel programming languages and hybrid solutions. This incremental approach is too conservative, often resulting in very complex code. This paper describes the design features, the programming constructs, and the runtime mechanisms of the Data Centric programming model for Exascale systems (DCEx). DCEx is based on structuring applications into data-parallel blocks. Blocks are units of shared-and distributed-memory parallel computation, communication, and migration in the memory/storage hierarchy. Blocks and their message queues are mapped onto processes and placed in memory/storage by the DCEx runtime. Those data-parallel blocks are orchestrated by using distributed parallel patterns that simplify the development cost. DCEx aims to reach the convergence of traditional HPC programming models, mainly based on MPI, with the emerging technologies based on the data intensive paradigms. To demonstrate the potential of DCEx, we carried out an experimental evaluation developing a real-world diffusion-weighted magnetic resonance imaging data processing application in a neuroimaging research context.

关键词： Big Data HPC convergence programming model PGAS parallel patterns

来源：评论

学校读者我要写书评

暂无评论

A programming model for Hybrid Collaborative Adaptive Systems

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING 2020年第1期8卷 6-19页

作者： Scekic, Ognjen Schiavinotto, Tommaso Videnov, Svetoslav Rovatsos, Michael Hong-Linh Truong Miorandi, Daniele Dustdar, Schahram TU Wien Distributed Syst Grp A-1040 Vienna Austria U Hopper I-38121 Trento Italy Univ Edinburgh Ctr Intelligent Syst & Their Applicat Edinburgh EH8 9YL Midlothian Scotland

Hybrid Diversity-aware Collective Adaptive Systems (HDA-CAS) are a new generation of socio-technical systems where both human and machine peers collectively participate in complex cognitive and physical tasks. These systems are characterized by the fundamental properties of hybridity and collectiveness, hiding from users the complexities associated with managing the collaboration and coordination of human-machine teams. The SmartSociety platform is a set of integrated software components that jointly provide a number of advanced HDA-CAS functionalities. As part of the CAS initiative, we have developed a programming model and Java APIs that make the use of those functionalities easy and accessible to application developers. In this paper we present the SmartSociety programming model elements, including the principal contributions - Collectives and Collective-based Tasks. We describe and discuss their functionality, implementation and runtime environment. Finally, we qualitatively evaluate the programming model and the language constructs with respect to the desired HDA-CAS properties.

关键词： programming Collaboration Computational modeling Software Adaptation models Adaptive systems Context Socio-technical systems collective adaptive systems crowdsourcing social computing programming model

来源：评论

学校读者我要写书评

暂无评论

INKS: a programming model to decouple algorithm from optimization in HPC codes

引用

JOURNAL OF SUPERCOMPUTING 2020年第6期76卷 4666-4681页

作者： Ejjaaouani, Ksander Aumage, Olivier Bigot, Julien Mehrenberger, Michel Murai, Hitoshi Nakao, Masahiro Sato, Mitsuhisa Univ Paris Saclay Maison Simulat CEA CNRSUniv Paris SudUVSQInria Gif Sur Yvette France INRIA LaBri Bordeaux France Univ Paris Saclay Maison Simulat CEA CNRSUniv Paris SudUVSQ Gif Sur Yvette France Univ Marseille Marseille France Riken CCS Kobe Hyogo Japan

Existing programming models tend to tightly interleave algorithm and optimization in HPC simulation codes. This requires scientists to become experts in both the simulated domain and the optimization process and makes the code difficult to maintain or port to new architectures. In this paper, we propose the INKS programming model that decouples these concerns with two distinct languages: INKSpia to express the simulation algorithm and INKSpso for optimizations. We define INKS (pia) and evaluate the feasibility of defining INKSpso with three test languages: INKSo/C++, INKSo/loop and INKSo/XMP. We evaluate the approach on synthetic benchmarks (NAS and heat equation) as well as on a more complex example (6D Vlasov-Poisson solver). Our evaluation demonstrates the soundness of the approach as it improves the separation of algorithmic and optimization concerns at no performance cost. We also identify a set of guidelines for the later full definition of the INKSpso language.

关键词： programming model Separation of concerns HPC DSL

来源：评论

学校读者我要写书评

暂无评论

programming model-based method for ranking objects from group decision making with interval-valued hesitant fuzzy preference relations

引用

APPLIED INTELLIGENCE 2019年第3期49卷 837-857页

作者： Zhang, Yuning Tang, Jie Meng, Fanyong Beijing Inst Technol Sch Management & Econ Beijing 100081 Peoples R China Cent S Univ Sch Business Changsha 410083 Hunan Peoples R China Nanjing Univ Informat Sci & Technol Sch Management & Econ Nanjing 210044 Jiangsu Peoples R China

Interval-valued hesitant fuzzy preference relations (IVHFPRs) are useful that allow decision makers to apply several intervals in [0, 1] to denote the uncertain hesitation preference. To derive the reasonable ranking order from group decision making with preference relations, two topics must be considered: consistency and consensus. This paper focuses on group decision making with IVHFPRs. First, a multiplicative consistency concept for IVHFPRs is defined. Then, programming models for judging the consistency of IVHFPRs are constructed. Meanwhile, an approach for deriving the interval fuzzy priority weight vector is introduced that adopts the consistency probability distribution as basis. Subsequently, this paper builds several multiplicative consistency-based programming models for estimating the missing values in incomplete IVHFPRs. A consensus index is introduced to measure the agreement degree between individual IVHFPRs, and a method for increasing the consensus level is presented. Finally, a multiplicative consistency-and-consensus-based group decision-making method with IVHFPRs is offered, and a practical decision-making problem is selected to show the application of the new method.

关键词： Group decision making IVHFPR Multiplicative consistency Consensus programming model

来源：评论

学校读者我要写书评

暂无评论

programming model-based group decision making with multiplicative linguistic intuitionistic fuzzy preference relations

引用

COMPUTERS & INDUSTRIAL ENGINEERING 2019年 136卷 212-224页

作者： Meng, Fanyong Tang, Jie Zhang, Yongliang Beijing Wuzi Univ Sch Informat Beijing 101149 Peoples R China Cent S Univ Sch Business Changsha 410083 Hunan Peoples R China

A new type of fuzzy sets called multiplicative linguistic intuitionistic fuzzy sets (MLIFSs) is introduced to denote the asymmetric qualitative preferred and non-preferred cognitions of decision makers. Considering the application of MLIFSs, an order relationship is offered and several operational laws are defined. Then, multiplicative linguistic intuitionistic fuzzy preference relations (MLIFPRs) are introduced. A consistency concept is defined and its several properties are researched. A programming model to judge the consistency of MLIFPRs is constructed, and an approach to derive consistent MLIFPRs is provided. Meanwhile, a programming model to ascertain missing values in incomplete MLIFPRs is built. A consensus index is defined to measure consensus and a programming model to improve the consensus level is constructed, which ensures the consistency of individual MLIFPRs. Furthermore, a programing model to determine the weights of the decision makers is established. On the basis of consistency and consensus analysis, an approach to group decision making with MLIFPRs is developed. Finally, an example about evaluating washing machine is selected to show the application of the new method.

关键词： Group decision making programming model MLIFPR Consistency and consensus

来源：评论

学校读者我要写书评

暂无评论

programming model and implementation mechanism for ternary optical computer

引用

OPTICS COMMUNICATIONS 2018年 428卷 26-34页

作者： Zhang Sulan Peng Junjie Shen Yunfu Wang Xianchao Shanghai Univ Sch Comp Engn & Sci Shanghai 200444 Peoples R China Fuyang Normal Univ Sch Math & Stat Fuyang 23600 Peoples R China

In order to make use of the advantages of optical computing and promote the application and popularization of Ternary Optical Computer (TOC), this paper proposes a new computing-data model, presents the corresponding computing-data file in detail, and gives the implementation method of programming application with the model files. It tries to set up a bridge between users and TOC. Experimental results show that the model presented in this paper is correct and the implementation method is feasible. The model can better simplify TOC's application process, and allow users to apply TOC to carry out various computational applications without needing to understand any details of optical implementation.

关键词： Computing-data model Computing-data file programming model Ternary optical computer Ternary optical processor

来源：评论

学校读者我要写书评

暂无评论

InKS, a programming model to Decouple Performance from Algorithm in HPC Codes 24th

InKS, a Programming Model to Decouple Performance from Algor...

引用

International European Conference on Parallel and Distributed Computing (Euro-Par)

作者： Ejjaaouani, Ksander Aumage, Olivier Bigot, Julien Mehrenberger, Michel Murai, Hitoshi Nakao, Masahiro Sato, Mitsuhisa UVSQ Univ Paris Saclay Univ Paris Sud Maison Simulat CEACNRS Gif Sur Yvette France INRIA Nancy France INRIA Bordeaux France LaBRI Bordeaux France Univ Strasbourg IRMA Strasbourg France Riken AICS Kobe Hyogo Japan

ISBN: (纸本)9783030105495;9783030105488

Existing programming models tend to tightly interleave algorithm and optimization in HPC simulation codes. This requires scientists to become experts in both the simulated domain and the optimization process and makes the code difficult to maintain and port to new architectures. This paper proposes the InKS programming model that decouples these two concerns with distinct languages for each. The simulation algorithm is expressed in the InKSpia language with no concern for machine-specific optimizations. Optimizations are expressed using both a family of dedicated optimizations DSLs (InKSO) and plain C++. InKSO relies on the InKSpia source to assist developers with common optimizations while C++ is used for less common ones. Our evaluation demonstrates the soundness of the approach by using it on synthetic benchmarks and the Vlasov-Poisson equation. It shows that InKS offers separation of concerns at no performance cost.

关键词： programming model Separation of concerns HPC DSL

来源：评论

学校读者我要写书评

暂无评论

An Improved programming model for Thread-Level Speculation 17

An Improved Programming Model for Thread-Level Speculation

引用

IEEE Int Conf on Parallel and Distributed Processing with Applications, Big Data and Cloud Computing, Sustainable Computing and Communications, Social Computing and Networking (ISPA/BDCloud/SocialCom/SustainCom)

作者： Liu, Bin Yang, Hao Li, Yuancheng Li, Yuxiang Niu, Dangdang Lv, Zhiming Northwest A&F Univ Coll Informat Engn Yangling Shaanxi Peoples R China Xian Univ Sci & Technol Sch Comp Sci & Technol Xian Peoples R China Henan Univ Sci & Technol Sch Informat Technol Luoyang Peoples R China

ISBN: (纸本)9781728143286

Thread-Level speculation (TLS) is an automatic parallel technique, which partitions sequential programs with data dependencies into multithreads to be executed in parallel on the multi-core platform, and it is used to aggressively exploit the parallelism of sequential programs at runtime. However, the existing implementations of the software TLS models always incur extra and high overhead, even could be revoking performance benefits from TLS when frequent data dependencies are violated. Based on the POSIX thread library, which has a high portability, this paper proposes an improved TLS programming model to exploit potential parallelism of the sequential programs. Firstly, the execution model of the proposed programming model adopts the optimized main-assistance-work thread mechanism to improve overall operational efficiency. And then a new data structure is proposed and designed to implement a thread-level speculation mechanism for solving the thread version copy. Using the benchmarks, experimental results show that the proposed TLS programming model can acquire a significant increase in speedup on a multi-core platform and the proposed TLS programming model has excellent performance benefits. Compared with the traditional TLS programming model, the improved TLS model can achieve a speedup of 2.87 to 12.22 which delivers a 9.70% performance improvement on average.

关键词： Thread-Level Speculation Multi-cores programming model Assistance Thread Pthreads

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：