检索结果-内蒙古大学图书馆

42nd Annual International Conference on Parallel Processing (ICPP)

作者： Rinke, Sebastian Prabhakaran, Suraj Wolf, Felix German Res Sch Simulat Sci D-52062 Aachen Germany

ISBN: (纸本)9780769551173

The integration of accelerators into cluster systems is currently one of the architectural trends in high performance computing. Usually, those accelerators are manycore compute devices which are directly connected to individual cluster nodes via PCI Express. Recent advances of accelerators, however, do not require a host CPU anymore and now even enable their integration as self-contained nodes that are able to MPI-communicate over their own network interface. This approach offers new opportunities for application developers, as compute kernels can now span multiple communicating accelerators to better account for larger MPI-based code regions with the potential for massive node-level parallelism. However, it also raises the question of how to program such an environment. An instance of this novel cluster architecture is the DEEP cluster system currently under development. Based on this hardware concept, we investigate the MPI_Comm_spawn process creation mechanism for offloading MPI-based distributed memory compute kernels onto multiple network-attached accelerators. We identify limitations of MPI_Comm_spawn and present an offloading mechanism which results in only a fraction of the overhead of a pure MPI_Comm_spawn solution.

关键词： MPI_Comm_spawn computation offloading network-attached accelerators Intel Xeon Phi DEEP

来源：评论

学校读者我要写书评

暂无评论

Towards computational offloading in Mobile Device Clouds

Towards Computational Offloading in Mobile Device Clouds

引用

5th IEEE International Conference on Cloud Computing Technology and Science (IEEE CloudCom)

作者： Mtibaa, Abderrahmen Harras, Khaled A. Fahim, Afnan Texas A&M Univ ECEN Dept Doha Qatar Carnegie Mellon Univ Sch Comp Sci Doha Qatar

ISBN: (纸本)9781479915484

Many mobile applications overcome their device limitations in computational, energy, or data resources by offloading computations to the cloud. In this paper, we consider environments in which computational offloading occurs amongst a set of mobile devices. We call such an environment a mobile device cloud (MDC). In this work, we first highlight the gain in computation time and energy consumption that can be achieved by offloading tasks to nearby devices within an MDC compared to a cloud. We then propose and implement an MDC platform that enables the creation and assessment of various offloading algorithms in MDCs. This platform consists of an Android application deployable across MDC devices, and a test bed to measure power being consumed by a mobile device. We utilize this platform to carry out various offloading experiments on an MDC testbed from which we gain interesting insights into the potential for MDC offloading. Results from these experiments show up to 50% gain in time and 26% gain in energy. Finally, we address the offloadee selection problem in MDCs by proposing several social-based algorithms. The potential promise of this approach is shown by evaluating these algorithms using real data sets that include contact traces and social information of mobile devices in a conference setting.

关键词： computation offloading Mobile Device Cloud Opportunistic Communication Opportunistic Computing

来源：评论

学校读者我要写书评

暂无评论

Smartphone Energizer: Extending Smartphone's Battery Life with Smart offloading

Smartphone Energizer: Extending Smartphone's Battery Life wi...

引用

9th International Wireless Communications and Mobile Computing Conference (IWCMC)

作者： Khairy, Ayat Ammar, Hany H. Bahgat, Reem Cairo Univ Fac Comp & Informat Dept Comp Sci Cairo Egypt West Virginia Univ Lane Comp Sci & Elect Engn Dept Morgantown WV 26506 USA

ISBN: (纸本)9781467324809

This paper presents "Smartphone Energizer", a novel technique for context-aware computation offloading for smartphones. Previous techniques to the offloading problem were based on a narrow set of contextual information, which made the amount of energy saving varies unexpectedly based on the context in which the application is running. Smartphone Energizer uses the benefit of supervised learning with a rich set of contextual information such as application, device, network, and user characteristics to optimize both the energy consumption and execution time of Smartphone's applications in a variety of contextual situations. In our evaluation, we show that Smartphone Energizer predicts both energy consumption and execution time in different contexts with error less than 9%, which in turn helped in taking the right offloading decision and saving energy by 40% to 56% and execution time by 43% to 58%.

关键词： computation offloading context-awareness application partitioning Smartphones energy efficiency

来源：评论

学校读者我要写书评

暂无评论

Towards an Energy Efficient Code Generator for Mobile Phones

Towards an Energy Efficient Code Generator for Mobile Phones

引用

4th IEEE International Conference on Cognitive Infocommunications (CogInfoCom)

作者： Fekete, Krisztian Csorba, Kristof Vajk, Tamas Forstner, Bertalan Pandi, Krisztian Budapest Univ Technol & Econ Dept Automat & Appl Informat Budapest Hungary

ISBN: (纸本)9781479915439;9781479915460

Using a smartphone become the part of our everyday life in the last few years. These devices can help us in many areas of life (sport, job, weather etc.), but sometimes can be also very annoying because of the battery life time. That is why it is very important to find solutions, which can reduce the energy consumption of the smartphones. One possible method is the " computation offloading" where a part of the processes are executed on a remote device (e. g. in the cloud). A lot of example has already shown, that computation offloading can reduce the energy usage of the mobile devices. However, the amount of energy saving may differ, as the decision making of the offloading process can be controlled with several techniques. Our decision making theory is based on the scheduling theory. In this paper, we are going to introduce a new system called ECGM (Energy efficient Code Generator for Mobile phones). ECGM decides automatically at compile time, which task should run on the smartphone and which task can be offloaded. The benefit of our system will be demonstrated through measurements based on our energy-efficiency scheduling technique.

关键词： computation offloading Energy Consumption Cloud Mobile Computing Job Scheduling Tail energy Compiler Automation Microsoft Roslyn

来源：评论

学校读者我要写书评

暂无评论

MOCLO: A cloud framework for mobile devices

MOCLO: A cloud framework for mobile devices

引用

2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing, GreenCom-iThings-CPSCom 2013

作者： Gangil, Akash Dhurandher, Sanjay K. Obaidat, Mohammad S. Singh, Vijit Bhatia, Sushant Department of Computer Engineering Netaji Subhas Institute of Technology University of Delhi New Delhi India Division of Information Technology Netaji Subhas Institute of Technology University of Delhi New Delhi India Department of Computer Science and Software Engineering Monmouth University NJ United States

ISBN: (纸本)9780769550466

With the advent of smartphones, mobile-devices have evolved as powerful mini-computers. However, they still suffer from two intrinsic shortcomings, namely limited battery-power and inferior processors compared to their laptop counterparts. Several approaches based on cloud-computing have been suggested in the past to augment the performance of mobile devices and overcome these shortcomings. The utility of these approaches is demarcated by the maximum bandwidth and minimum latency that could be achieved in a cloud over the Internet. In this paper we propose MOCLO, a framework that operates upon locally-installed cloud called MOCLETS which is free from majority of shortcomings present in other Internet based cyber-foraging models. The paper also outlines the optimal architecture of mobile-applications to gain maximum benefit from MOCLO. A proof-of-concept prototype of MOCLO has also been implemented. Results have shown significant decrease in battery consumption and processing time of mobile applications leveraging MOCLO APIs. We also outline the significant cyber foraging frameworks proposed so far and where MOCLO stands in comparison to them. Towards the end we discuss the limitations of MOCLO and scalability issues and consider possible solutions for the same. © 2013 IEEE.

关键词： computation offloading

来源：评论

学校读者我要写书评

暂无评论

Empowering mobile service provisioning through cloud assistance

Empowering mobile service provisioning through cloud assista...

引用

2013 IEEE/ACM 6th International Conference on Utility and Cloud Computing, UCC 2013

作者： Elgazzar, Khalid Martin, Patrick Hassanein, Hossam S. School of Computing Queen's University Canada

ISBN: (纸本)9780769551524

The use of mobile devices as data service providers is on the rise. Mobile devices feature a large set of distinct characteristics that qualify them to be the most convenient computing platform for online services, both as consumers and providers. Mobile devices can take advantage of their mobility to provide location-based services and their association to a specific user to customize service offerings to fit personal preferences and current conditions. Howev, the increasing resource demands of mobile services and the inherent constraints of mobile devices limit the quality and type of functionality that can be offered, preventing mobile devices from exploiting their full potential as reliable data providers. Cloud computing offers mobile devices the opportunity to run resource-intensive tasks through computation offloading. The offloading decision is a tradeoff between data transfer and latency improvement to the benefit of alleviating the burden on mobile resource while improving the overall performance of service provisioning. This paper presents a framework for cloud-assisted mobile service provisioning, aimed at offering an augmented environment to resource-constrained mobile providers in order to deliver reliable services. The framework supports dynamic offloading based on the resource status at the mobile side and current network condition as well as user-defined energy constraints. It also enables the mobile provider to delegate the cloud to forward the service response directly to the user, given that no further processing is required by the provider. Performance evaluation shows up to 6x latency improvement for computational-intensive services that do not require large data transfer. © 2013 IEEE.

关键词： computation offloading

来源：评论

学校读者我要写书评

暂无评论

Making the case for computational offloading in mobile device clouds 13

Making the case for computational offloading in mobile devic...

引用

19th Annual International Conference on Mobile Computing and Networking, MobiCom 2013

作者： Fahim, Afnan Mtibaa, Abderrahmen Harras, Khaled A. School of Computer Science Carnegie Mellon University Doha Qatar ECE Department Texas A and M University Doha United States

ISBN: (纸本)9781450319997

In this paper, we consider an environment in which compu- Tational offloading is adopted amongst mobile devices. We call such an environment a mobile device cloud (MDC). In this work, we highlight via emulation, experimenation and real measurements, the potential gain in computation time and energy consumption that can be achieved by offload- ing tasks within an MDC. We also propose and develop an experimental platform to enable researchers create and ex- periment with novel offloading algorithms in MDCs. © 2013 by the Association for Computing Machinery, Inc.

关键词： computation offloading

来源：评论

学校读者我要写书评

暂无评论

ON THE EFFICACY OF computation offloading DECISION-MAKING STRATEGIES

引用

INTERNATIONAL JOURNAL OF HIGH PERFORMANCE COMPUTING APPLICATIONS 2008年第4期22卷 460-479页

作者： Gurun, Selim Wolski, Rich Krintz, Chandra Nurmi, Dan Univ Calif Santa Barbara Dept Comp Sci Santa Barbara CA 93106 USA

We present a framework for making computation offloading decisions in computational grid settings in which schedulers determine when to move parts of a computation to more capable resources to improve performance. Such schedulers must predict when an offloaded computation will outperform one that is local by forecasting the local cost (execution time for computing locally) and remote cost (execution time for computing remotely and transmission time for the input/output of the computation to/from the remote system). Typically, this decision amounts to predicting the bandwidth between the local and remote systems to estimate these costs. Our framework unifies such decision models by formulating the problem as a statistical decision problem that can either be treated "classically" or using a Bayesian approach. Using an implementation of this framework, we evaluate the efficacy of a number of different decision strategies (several of which have been employed by previous systems). Our results indicate that a Bayesian approach employing automatic change-point detection when estimating the prior distribution is the best performing approach.

关键词： computational grid computation offloading remote execution network prediction RPC

来源：评论

学校读者我要写书评

暂无评论

offloading of Media Transcoding for High-Quality Multimedia Services

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 2012年第2期58卷 691-699页

作者： Kim, Seung Hun Kim, Keunsoo Lee, Changmin Ro, Won W. Yonsei Univ Embedded Syst & Comp Architecture Lab Sch Elect & Elect Engn Seoul 120749 South Korea

The demand for high-quality multimedia service in handheld devices has increased with the advance of consumer electronics technologies. However, there are inherent performance limits in such mobile devices due to low processing power, requirements for low power consumption, and limited storage capacity. These factors have obviously prevented high-performance multimedia services from being feasible. In addition, most of the mobile devices support distinct multimedia data formats for video and audio codecs, bit rate, and screen size. Therefore, streaming media or receiving data from other users' devices essentially requires additional data translation along with a transcoding operation. However, most of the processors that are used in mobile devices or even in personal media servers cannot provide sufficient processing power for transcoding. To address the problem, this paper proposes an efficient offloading method for media transcoding can be applied in current commercial products. The system is designed on top of Java and provides sufficient transcoding performance for media streaming with low-end processors.

关键词： computation offloading transcoding media streaming mobile devices low-power consumption

来源：评论

学校读者我要写书评

暂无评论

Poster Abstract : Mobile Application Partitioning for Improving Energy Efficient 12

Poster Abstract : Mobile Application Partitioning for Improv...

引用

3rd IEEE/ACM International Conference on Cyber-Physical Systems (ICCPS)

作者： Niu, Jianwei Gao, Yuhang Niu, Ruifang Beihang Univ Sch Comp Sci & Engn Beijing Peoples R China Henan Univ Sci & Technol Sch Elect & Informat Engn Luoyang Peoples R China

ISBN: (纸本)9780769546957

For resource-constraint devices, it is possible to save energy of mobile devices by offloading its computation to a remote server. Traditionally, the application partitioning is the key element of computation offloading. However, making a good partitioning needs to tradeoff the communication cost and computational cost. To address this problem, we propose Energy-Aware Dynamic Application Partitioning (EDAP). It first constructs the Weight Object Relation Graph (WORG) for a given java application program by using the static analysis and dynamic profiling techniques. Then, we transform the WORG into Energy Graph (EG) to minimize the energy consumption. By using the network flow theory, the optimization problem of energy consumption is transformed into the optimal bipartition problem of a flow network which can be partitioned by the max-flow min-cut algorithm.

关键词： application partitioning computation offloading mobile device energy saving

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：