3220 Slices
Medium 9781601323262

Miss-rate Based Dynamic Phase Tracking on Multicore Processors

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

Int'l Conf. Par. and Dist. Proc. Tech. and Appl. | PDPTA'14 |

341

Miss-rate Based Dynamic Phase Tracking on

Multicore Processors

Sourav Dutta

Nazeih M. Botros

Department of Electrical and

Computer Engineering

Southern Illinois University

Carbondale, Illinois 62901–6899

Email: sourav@siu.edu

Department of Electrical and

Computer Engineering

Southern Illinois University

Carbondale, Illinois 62901–6899

Email: botrosn@siu.edu

Corresponding Author

Abstract—In this paper we propose a method to detect phases of an application with similar L2 cache miss behavior. Our proposed algorithm can predict per phase cache miss rate and able to detect stable phase transitions dynamically. In order to predict the L2 cache miss behavior of an application for different phases,

We exploit the fractal behavior of cache and propose a miss rate adaption technique which allows to detect phase transitions. In this approach we use performance counters already available in modern processors. Adaptive behavior of the presented work makes it useful for real time applications. We performed different applications from modern benchmark to evaluate our proposed method. We use sniper multi-core simulator to model a multicore out-of-order processor resembling the modern Intel Nehalem processor. Experimental results show that the proposed method has about 95% accuracy with minimum 2.1% prediction error and significantly low overhead.

See All Chapters
Medium 9781601323545

SESSION Novel Studies

Hamid Arabnia, Fernando G. Tinetti, Quoc-Nam Tran CSREA Press PDF

Int'l Conf. Bioinformatics and Computational Biology | BIOCOMP'16 |

SESSION

NOVEL STUDIES

Chair(s)

TBA

ISBN: 1-60132-428-6, CSREA Press ©

115

116

Int'l Conf. Bioinformatics and Computational Biology | BIOCOMP'16 |

ISBN: 1-60132-428-6, CSREA Press ©

Int'l Conf. Bioinformatics and Computational Biology | BIOCOMP'16 |

117

Recurrent Breast Cancer Treatment via

Rapid Pain-Guided Experimentation

Jane Eyre1, Steve Richfield IEEE # 413447142

1

Owner of NormalBodyTemperature.co.uk TudorJane@gmail.com

2

Owner of FixLowBodyTemp.com

Steve.Richfield@gmail.com

Abstract – Recurrent breast cancers present a special challenge. Spawning from mutated surviving cancer cells or oncogenic cysts, they tend to be aggressive and malignant, and many have mutated to circumvent measures that killed an earlier tumor, so survival rates are negligible.

There is a brief window of time, starting when tumors first become painful, and ending when they have grown enough to do sufficient damage to surrounding tissues so that pain cannot be quickly stopped. When tumors are exquisitely sensitive to their internal pressure, it is possible to quickly test prospective adjuvant therapies and receive immediate feedback as to their efficacy in the form of prompt reduction or elimination of pain.

See All Chapters
Medium 9781601323262

A Novel Method to Minimize Side Effects by Cache Contention on the Inclusive Multi-Core Shared Cache

Hamid R. Arabnia, Lou D'Alotto, Hiroshi Ishii, Minoru Ito, Kazuki Joe, Hiroaki Nishikawa, Georgios Sirakoulis, William Spataro, Giuseppe A. Trunfio, George A. Gravvanis, George Jandieri, Ashu M. G. Solo, Fernando G. Tinetti CSREA Press PDF

A Novel Method to Minimize Side Effects by Cache

Contention on the Inclusive Multi-Core Shared Cache

Hyo-Joong Suh1

School of Computer Science and Information Engineering,

The Catholic University of Korea, Bucheon-si, Gyeonggi-do, Korea

1

Abstract - Cache contention is annoying issue of the multicore processors with private L1 cache and inclusive shared L2 cache. If core P1 runs with small working set which fits in its

L1 cache while core P2 runs cache consuming process, loaded blocks in the shared L2 cache by P1 are extinguished by frequent access from P2. Furthermore, it incurs an inevitable invalidate to the L1 cache of the core P1 by the multi-level cache inclusion property. This study focused on this problem, and solves by access grouping with extension of cache status. Simulation results show the proposed method minimizes the side effects by the cache contention on the inclusive shared cache.

workloads are dispatched on the each core with shared L2 cache, lightweight core is disturbed by invalidation requests to maintain the inclusion property between caches. Fig. 2 shows the cache contention [2] and invalidation by inclusion which this study addressed to resolve.

See All Chapters
Medium 9781601322494

A Mechanism to Prevent Side Channel Attacks in Cloud Computing Environments

Hamid R. Arabnia; Victor A. Clincy; Leonidas Deligiannidis; Ashu M. G. Solo; and Fernando G. Tinetti (Editors) Mercury Learning and Information PDF

Int'l Conf. Internet Computing and Big Data | ICOMP'13 |

229

A Mechanism to Prevent Side Channel Attacks in Cloud Computing

Environments

Tzong-An Su

Fenh Chia University

Taichung, Taiwan

Abstract - Cloud computing provides

the benefits of scalability, agility, reliability, security, performance, and maintenance to enterprises and has emerged to become a reality. While cloud computing brings in many benefits it also introduces several security issues. One of the most serious security issues is the side channel attacks which are attacks based on information gained from the physical implementation of a machine. In this paper, we aim at providing a platform to prevent various types of side channel attacks, e.g., cache-based and timing attacks.

Keywords – Side Channel Attack,

Cloud Computing, Hardware

Virtualization, Virtual Machine

1

1.1

Introduction

Background

Cloud Computing has been a buzzword for the pass few years. The concept combined technologies in networking and computing to provide various services to individuals and organizations. Services like SaaS (Software as a Service), Uaas

See All Chapters
Medium 9781601323248

An Affine Shape Constraint for Geometric Active Contours

Hamid R. Arabnia, Leonidas Deligiannidis Joan Lu, Fernando G. Tinetti, Jane You, George Jandieri, Gerald Schaefer, and Ashu M.G. Solo CSREA Press PDF

Int'l Conf. IP, Comp. Vision, and Pattern Recognition | IPCV'14 |

343

AN AFFINE SHAPE CONSTRAINT FOR GEOMETRIC ACTIVE CONTOURS

Mohamed Amine Mezghich, Mawaheb Saidani, Slim M’Hiri and Faouzi Ghorbel

GRIFT Research Group, CRISTAL Laboratory

´

Ecole Nationale des Sciences de l’Informatique, ENSI

Campus Universitaire de la Manouba, 2010 Manouba, Tunisia ma.mezghich@cristal.rnu.tn, {mawaheb.saidani,slim.mhiri,faouzi.ghorbel}@ensi.rnu.tn

ABSTRACT

We intend to present in this research a new method to incorporate affine invariant geometric shape prior into a level set based active contours for robust segmentation of partially occluded object. The proposed shape constraint is defined after an affine shape alignment of the active contour and a reference shape. In order to generalize our work to the multi-references case, we present a set of complete and stable invariant shape descriptors computed using Fourier transform of the contours to choose the most suitable reference according to the evolving contour. Experiments on synthetic and real data show the ability of the proposed approach to constrain an evolving curve towards a target shapes that may be occluded and cluttered under geometric transformations.

See All Chapters

See All Slices