MIOS:面向大规模CCNUMA系统的多实例操作系统  

MIOS:A Scalable Multi-Instance OS for Large Scale CCNUMA System

在线阅读下载全文

作  者:卢凯[1] 迟万庆[1] 高颖慧[2] 冯华[1] 

机构地区:[1]国防科学技术大学计算机学院,长沙410073 [2]国防科学技术大学自动目标识别实验室,长沙410073

出  处:《计算机研究与发展》2011年第9期1693-1703,共11页Journal of Computer Research and Development

基  金:霍英东青年教师基金项目(111072)

摘  要:MIOS是一个面向大规模CCNUMA系统设计的新型高可扩展操作系统.MIOS创新地采用了多实例内核结构,每个内核实例执行相同代码,分别独立运行和管理一个处理器,多核间通过分布存储管理构成高可扩展的一致性系统映像空间,支持弱共享进程、线程并行模型.MIOS针对大规模CCNUMA系统特点和高性能并行科学计算应用的需求,采用了显式共享数据分布、层次式任务调度、自适应任务间通信以及寄存器锁等优化.在大规模CCNUMA体系结构的银河深度并行计算机上的测试表明,MIOS对MPI应用具有同传统操作系统类似的性能,并可以有效支持2048处理器规模的OMP应用高效运行,具有良好的系统可扩展性.MIOS is a scalable operating system designed for large scale CCNUMA system. It introduces multi-instance kernel structure. In MIOS, each instance of OS kernel executes the same code, but runs on a node of the CCNUMA machine and manages its resources respectively. The MIOS provides a single system image running environment for all nodes of CCNUMA system, supporting process and thread task model. Aiming at the features of CCNUMA system and the requirements of scientific computing applications, the MIOS provides several optimizations, including weak shared thread model, cascaded task scheduling, adaptive communication between tasks and register-based lock. We have implemented MIOS on our Galaxy parallel computer system, a large scale CCNUMA system including 2048 processors. The evaluations on Galaxy system, including micro-benchmarks and real parallel applications, show that MIOS can provide comparable performance with a conventional OS for MPI applications. For OMP applications, the MIOS also can provide a good performance speedup on the large scale CCNUMA system with 2048 processors. The structure of MIOS can also provide experiences for designing operating system on many-core processor.

关 键 词:CCNUMA 多核系统 并行执行模型 存储管理 操作系统 

分 类 号:TP316[自动化与计算机技术—计算机软件与理论]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象