19 Chapters
Medium 9781601323187

Author Index

Hamid R. Arabnia, George A. Gravvanis, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF
Medium 9781601323187

Cloud Computing Benchmarking: A Survey

Hamid R. Arabnia, George A. Gravvanis, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Grid & Cloud Computing and Applications | GCA'14 |

15

Cloud Computing Benchmarking: A Survey

C.Vazquez, R. Krishnan, and E. John

Department of Electrical and Computer Engineering, The University of Texas at San Antonio,

San Antonio, Texas, U.S.A

Abstract – Cloud computing gives service-oriented access to computing, storage and networking resource. Often, these resources are virtualized. The prospect of being able to scale computing resources to meet user demand has clearly caught the attention of developers and organizational IT leaders over the recent years. Considering the number of cloud computing providers and the different services each provider offers, cloud users need benchmark information that specifically addresses the unique properties of the cloud computing environment such as dynamic scaling. This paper compares five prominent tools (CloudCmp, CloudStone, HiBench, YCSB, and CloudSuite) that present workloads and/or methods for quantitatively comparing cloud computing offerings.

See All Chapters
Medium 9781601323187

Session - Cloud Computing, Analysis, and Performance Evaluation + Big Data

Hamid R. Arabnia, George A. Gravvanis, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF
Medium 9781601323187

Analysis of ICmetrics features requirements in Cloud environment

Hamid R. Arabnia, George A. Gravvanis, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF

10

Int'l Conf. Grid & Cloud Computing and Applications | GCA'14 |

Analysis of ICmetrics features requirements in Cloud environment

%LQ�

School of Engineering and Digital Arts

University of Kent

Canterbury, UK by30@kent.ac.uk

M.Haciosman, Gareth Howells

School of Engineering and Digital Arts

University of Kent

Canterbury, UK mh521@kent.ac.uk , W.G.J.Howells@kent.ac.uk

Abstract—As web-server spoofing is increasing, we investigate a novel technology termed ICmetrics, used to identify fraud for given web servers based on measurable quantities/features. The novel concept ICmetrics is used to detect spoof websites with the advantages of a higher level of security with increased speed and template free encryption. ICmetrics technology is based on extracting features from digital systems’ operation that may be integrated together to generate unique identifiers for each of the systems or create unique profiles that describe the systems’ actual behavior. Ideally, the nature of the features should be identical for all of the systems considered, while the values of these features should allow for unique identification of each of the system servers. This paper looks at the properties of the several behaviors as a potential ICmetrics features, and explores properties which affects the stability of the system’s performance. We conclude three requirements for ICmetrics system.

See All Chapters
Medium 9781601323187

Research and Implementation of a New Cloud Server

Hamid R. Arabnia, George A. Gravvanis, Ashu M. G. Solo, and Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Grid & Cloud Computing and Applications | GCA'14 |

79

Research and Implementation of a New Cloud Server

1, 3, 5

Hua Nie1, Xiaojun Yang2, Chaoqun Sha3, Yanping Gao4, and Keping Long5

School of Computer and Communication Engineering, University of Science and Technology Beijing.

Beijing, China

2

Dawning Information Industry Co., Ltd. Beijing, China

4

Institute of Computing Technology, Chinese Academy of Sciences. Beijing, China

4

Loongson Technology Co., Ltd. Beijing, China

Abstract - Instead of all using TOR schemes, an approach building a cloud server on top of a high-performance fabric is presented in this paper. The advantage is to provide both high performance/cost and performance/Watt compared with the existing method. A FPGA-based system controller integrated with shared networking, shared storage, and interconnect fabric controller is designed and implemented to interconnect a set of lightweight server processors for building a highdensity server. All the processors can share the networking and storage resources through an inter-system interconnect fabric. For the 64-processor prototyping system, the evaluating results show the cloud server not only keeps some traditional cluster advantages such as OS compatibility, but also achieves the better scalability, high performance/cost and high performance/Watt for workloads.

See All Chapters

See All Chapters