The 2009 IEEE International Workshop on Multi-Core Software Systems

in conjunction with The 15th International Conference on Parallel and Distributed Systems


Due to power and processor complexity limit of single core, many manufactures have switched to multi-core or many-core processors. Compared with single-core design, multi-core or many-core processors can accommodate many computing units in one chip and exploit thread level parallel (TLP) to improve the performance of applications. Trends seen from current multi-core chip suggest that cores in one chip are still increasing and we are stepping into a new era where multi-core or many-core computing will be ubiquitous in personal computers, servers and embedded platforms. As the middle layer between applications and hardware, whether system software can exploit the powerful hardware resource effectively and accelerates applications is a valuable research direction.


Topics of interest


-Topics of interest on Multi-Core Software systems include, but are not limited to:

 > programming languages and models

 > algorithms and applications

 > system software models

 > compiler technology

 > runtime systems

 > operating system

 > workload characterization

 > performance modeling and evaluation


Important Dates


Paper Submission Due: 31 July 2009

Acceptance Notification: 1 September 2009

Final Manuscript Due: 1 October 2009


Submission Guidance


Please email your manuscript in PDF to with the subject of "MCSS submission - [AuthorName]". All papers submitted should not be published or currently under review elsewhere. Papers should be in IEEE CS format. All papers will be peer reviewed by PC members. Submitting a paper to the workshop means that if the paper is accepted, at least one author is required to register and to present the paper.


Publication of Papers


Papers accepted for ICPADS 2009 workshops will be published by the IEEE Computer Society and distributed to the participants. All accepted papers will be included in IEEE Xplore and indexed by EI.


General Co-Chairs


Weimin Zheng, Tsinghua University, China

Xiangke Liao, National University of Defence Technology, China.


Program Chair


Yu Chen , Tsinghua University, China


Program Committee(TBC)


Wenguang Chen, Tsinghua University, China

Laurence T. Yang, St. Francis Xavier University,Canada

Qinbo Wu, National University of Defence Technology, China.

Eddie Dong, Intel

Qingguo Zhou, Lan zhou University, China

Aimin Pan, Microsoft Research Asia

Xiaoge Wang, Tsinghua University, China

Li Zhao, Intel

Zhihui Du, Tsinghua University, China

Xiaofei Liao, Huazhong University of Science and Technology

Huadong Dai, National University of Defence Technology, China

Haibo Chen, Fudan University, China


Contact Person


Yu Chen

Tsinghua University, China

T. +8610 62784240


<yuchen [at]>

<chyyuu [at]>



Workshop Program


Time14:00PM~18:30PM, Tuesday, December 11, 2009

LocationRoom 4


(Time: 14:00PM-14:05PM  )

Welcome Message from Workshop Chairs

General ChairWeimin Zheng, Tsinghua University, China

General ChairXiangke Liao, National University of Defence Technology, China.

Program Chair: Yu Chen, Tsinghua University, China


(Time: 14:05PM-14:35PM  )

Adaptive Multi-versioning for OpenMP Parallelization via Machine Learning  

Xuan Chen, Shun Long


(Time: 14:35PM-15:05PM  )

Parallel Lexicographic Names Construction with CUDA

Weidong Sun, Zongmin Ma


(Time: 15:05PM-15:35PM  )

Count Sort for GPU Computing

Weidong Sun, Zongmin Ma


(Time: 15:35PM-16:05PM  )

Low Overhead Object Communication Scheme in CMP Implementation of Object-Oriented Programming

Licheng Xue, Feng Shi


(Time: 16:05PM-16:30PM  )

Coffee Break



(Time: 16:30PM-17:00PM  )

Challenges of Multi-core Computings Based on CBE and CUDA

Yingwen Song


(Time: 17:00PM-17:30PM  )

Solving 2D Nonlinear Unsteady Convection-Diffusion Equations on heterogenous platforms with multiple GPUs

Canqun Yang, Zhen Ge, Juan Chen, Feng Wang, Yunfei Du



(Time: 17:30PM-18:00PM  )

A Process Fusion Approach for MPI Performance Enhancement on Multi-core Systems

 Li Lu, Xiaoge Wang


(Time: 18:00PM-18:30PM  )

Domain-specific Pattern Matching based Automatic Parallelization:Demonstrated by 2-D Prestack Migration

Zhen Cao, Yuan Dong, Shengyuan Wang