基本信息
浏览量:11
职业迁徙
个人简介
Research Topics
Fault Tolerant HPC Systems
ULFM - User Level Failure Mitigation is a set of MPI interface extensions enabling Message Passing programs to restore MPI communication capabilies affected by process failures. It supports rebuilding communicators, RMA windows and I/O Files. No particular recovery model is imposed or favored, instead a set of versatile APIs is included that provides support for differente recovery styles. The application directs the recovery, so it can pay for the cost of repairing only the necessary MPI objects. The ULFM specification is a crucial infrastructure to enable the deployment of advanced, production quality fault toleant techniques; it is a versatile solution to improve the efficiency of novel and established fault tolerant techniques. Look at the flyer.
MPICH-V is a research effort with theoretical studies, experimental evaluations and pragmatic implementations aiming to provide a MPI implementation based on MPICH, featuring multiple fault tolerant protocols. MPICH-V provides automatic fault tolerant MPI library (i.e. a totaly unchanged application linked with the mpich-v library is a fault tolerant application).
Dataflow Execution Model for HPC
PaRSEC - Parallel Runtime Scheduling and Execution Controller - is a generic framework for architecture aware scheduling and management of micro-tasks on distributed many-core heterogeneous architectures. Applications we consider can be expressed as a Direct Acyclic Graph of tasks with labeled edges designating data dependencies. DAGs are represented in a compact problem-size independent format that can be queried on-demand to discover data dependencies in a totally distributed fashion. PaRSEC assigns computation threads to the cores, overlaps communications and computations and uses a dynamic, fully-distributed scheduler based on architectural features such as NUMA nodes and algorithmic features such as data reuse. The framework includes libraries, a runtime system, and development tools to help application developers tackle the difficult task of porting their applications to highly heterogeneous and diverse environment. PaRSEC is the underlying infrastructure for the DPLASMA distributed memory, tile algorithm based linear algebra package.
Fault Tolerant HPC Systems
ULFM - User Level Failure Mitigation is a set of MPI interface extensions enabling Message Passing programs to restore MPI communication capabilies affected by process failures. It supports rebuilding communicators, RMA windows and I/O Files. No particular recovery model is imposed or favored, instead a set of versatile APIs is included that provides support for differente recovery styles. The application directs the recovery, so it can pay for the cost of repairing only the necessary MPI objects. The ULFM specification is a crucial infrastructure to enable the deployment of advanced, production quality fault toleant techniques; it is a versatile solution to improve the efficiency of novel and established fault tolerant techniques. Look at the flyer.
MPICH-V is a research effort with theoretical studies, experimental evaluations and pragmatic implementations aiming to provide a MPI implementation based on MPICH, featuring multiple fault tolerant protocols. MPICH-V provides automatic fault tolerant MPI library (i.e. a totaly unchanged application linked with the mpich-v library is a fault tolerant application).
Dataflow Execution Model for HPC
PaRSEC - Parallel Runtime Scheduling and Execution Controller - is a generic framework for architecture aware scheduling and management of micro-tasks on distributed many-core heterogeneous architectures. Applications we consider can be expressed as a Direct Acyclic Graph of tasks with labeled edges designating data dependencies. DAGs are represented in a compact problem-size independent format that can be queried on-demand to discover data dependencies in a totally distributed fashion. PaRSEC assigns computation threads to the cores, overlaps communications and computations and uses a dynamic, fully-distributed scheduler based on architectural features such as NUMA nodes and algorithmic features such as data reuse. The framework includes libraries, a runtime system, and development tools to help application developers tackle the difficult task of porting their applications to highly heterogeneous and diverse environment. PaRSEC is the underlying infrastructure for the DPLASMA distributed memory, tile algorithm based linear algebra package.
研究兴趣
论文共 95 篇作者统计合作学者相似作者
按年份排序按引用量排序主题筛选期刊级别筛选合作者筛选合作机构筛选
时间
引用量
主题
期刊级别
合作者
合作机构
SC-W '23: Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysispp.44-50, (2023)
引用0浏览0EI引用
0
0
Int. J. Netw. Comput.no. 1 (2022): 26-46
HAL (Le Centre pour la Communication Scientifique Directe) (2022)
引用0浏览0引用
0
0
2022 IEEE/ACM 12th Workshop on Fault Tolerance for HPC at eXtreme Scale (FTXS)pp.36-46, (2022)
Future Generation Computer Systems (2020): 467-481
加载更多
作者统计
合作学者
合作机构
D-Core
- 合作者
- 学生
- 导师
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn