Dongxin Wen
Harbin Institute of Technology
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Dongxin Wen.
Journal of Systems Architecture | 2010
Dongxin Wen; Xiaoqun Wang; Xiaozong Yang
The transactional memory in multicore processors has been a major research area over past several years. Many transactional memory systems have been proposed to be used to solve the synchronization problem of multicore processors. Hardware transactional memory is one of the critical methods to speedup communications in multicore environment. In this paper, we give a review of the current hardware transactional memory systems for multicore processors. We take a top-down approach to characterizing and classifying various hardware transactional design issues and present a taxonomy of hardware transactional memory systems which is consist of the five fundamental design issues: version management, conflict detection, contention management, virtualization and nesting. Finally, we discussed the active research challenge: the relationship between transactional memory and Input/Output operations and system calls.
international conference on information engineering and computer science | 2009
Hongwei Liu; Xiaoqun Wang; Dongxin Wen; Xiaozong Yang
The transactional memory in multicore processors has been a major area of research over past ten years. Many transactional memory architectures have been proposed to solve the synchronization problem of multicore processors. Hardware transactional memory is one of the critical methods to speedup communications between many cores. We give a review of the current Hardware Transactional Memory systems for Multicore processors. Hardware transactional memory systems are classified into the following two categories: whether to support unbounded transactional memory and whether to support transactions nesting. Finally, we discuss two active research challenges: the relationship between transactional memory and Input/Output operations and Instruction Set ArchitectureThe transactional memory in multicore processors has been a very hot research area over past several years. Many transactional memory systems have been proposed to solve the synchronization problem of multicore processors. Software transactional memory is one of the critical methods to ease parallel programming and improve the scalability in the environment with many cores. In this study, software transactional memory systems for Multicore processors are reviewed according to the following aspects: transaction granularity, version management, conflict detection and synchronization. Finally, we discuss an active research challenge: whether strong isolation should be supported for the tradeoff between performance and semantics correctness in software transactional memory systems. KeywordsSoftware; Synchronization; Transactional Memory; Multicore processor; Parallel programming
international symposium on information and automation | 2010
Dongxin Wen; Xiaoqun Wang; Xiaozong Yang
In Transactional Memory (TM) for multicore systems, contention management refers to the mechanisms used to guarantee forward to avoid performance pathologies, and to promote throughput. The choice of contention management police impacts strongly on the performance of applications. In this paper, we study contention management policies for Hardware Transactional Memory (HTM). Although the results were obtained from a HTM framework, the conclusions and proposals are applicable to any type of TM: hardware, software or hybrid. We first characterize transactions’ behaviors on execution time.We find that significant time is wasted when few transactions forward progress due to the data conflicts among different transactions. To reduce this kind of waste time, we propose a novel contention management(CM) scheme that reduces waste time and guarantees forwarding. We evaluate our techniques using a detailed execution-driven simulator. The results show that our scheme can effectively minimize the amount of waste time and, consequently, decrease execution time than previous approaches for large transactions.
Archive | 2012
Decheng Zuo; Zhan Zhang; Qian Jun; Haiying Zhou; Dongxin Wen; Yi Feng; Baili Miao; Danyan Luo; Xiaozong Yang; Zhongqiang Cui; Hongwei Liu; Jian Dong
Archive | 2008
Gang Cui; Hongwei Liu; Decheng Zuo; Zhibo Wu; Xiaozong Yang; Jian Dong; Yanjun Shu; Baili Miao; Zhongchuan Fu; Yuanzhi Jiang; Lin Xiang; Zhan Zhang; Danyan Luo; Ling Wang; Dongxin Wen; Ziyang Que
Archive | 2007
Xiaozong Yang; Hongwei Liu; Zhibo Wu; Decheng Zuo; Gang Cui; Yanjun Shu; Jian Dong; Dongxin Wen; Ling Miao Bailixiang Linzhang
Archive | 2011
Hongwei Liu; Yanjun Shu; Jian Dong; Decheng Zuo; Xiaozong Yang; Zhibo Wu; Dongxin Wen; Zhan Zhang; Haiying Zhou; Danyan Luo; Baili Miao; Qian Jun
Archive | 2008
Gang Cui; Lianke Zhou; Liang Chen; Zhongchuan Fu; Xiaozong Yang; Zhibo Wu; Hongwei Liu; Decheng Zuo; Jian Dong; Yanjun Shu; Baili Miao; Lin Xiang; Zhan Zhang; Danyan Luo; Ling Wang; Dongxin Wen
Archive | 2007
Gang Cui; Hongwei Liu; Decheng Zuo; Xiaozong Yang; Zhibo Guan; Jian Dong; Yanjun Shu; Baili Miao; Lin Xiang; Zhan Zhang; Danyan Luo; Dongxin Wen; Ling Wang; Ziyang Que
Archive | 2006
Gang Cui; Xiaozong Yang; Hongwei Liu; Decheng Zuo; Feng Qu; Baili Miao; Jian Dong; Dongxin Wen; Lin Xiang; Zhan Zhang; Danyan Luo; Yanjun Shu; Ling Wang; Yuchang Mo; Ming Dai