摘要 :
In this project, we have focused on a range of techniques to overcome the performance bottleneck of current multiprocessor systems and to increase the single-node e-ciency. These eorts include the design and implementation of a to...
展开
In this project, we have focused on a range of techniques to overcome the performance bottleneck of current multiprocessor systems and to increase the single-node e-ciency. These eorts include the design and implementation of a toolset to analyze memory access patterns of applications, the exploration of regularity metrics and their use to classify code behavior, and a set of microbenchmarks to assess and quantify the performance of SMP memory systems. In addition, we explored the use of Dynamic Access Ordering (DAO) techniques in the realm of shared memory multiprocessors.
收起
摘要 :
Speculative execution is a method to increase instruction level parallelism which can be exploited by both super-scalar and VLIW architectures. The key to a successful general speculation strategy is a repair mechanism to handle m...
展开
Speculative execution is a method to increase instruction level parallelism which can be exploited by both super-scalar and VLIW architectures. The key to a successful general speculation strategy is a repair mechanism to handle mispredicted branches and accurate reporting of exceptions for speculated instructions. Multiple instruction rollback is a technique developed for recovery from transient processor failure. Many of the difficulties encountered during recovery from branch misprediction or from instruction re-execution due to exception in a speculative execution architecture are similar to those encountered during multiple instruction rollback. The applicability of a recently developed compiler-assisted multiple instruction rollback scheme to aid in speculative execution repair is investigated. Extensions to the compiler-assisted scheme to support branch and exception repair are presented along with performance measurements across ten application programs.
收起
摘要 :
Speculative execution is a method to increase instruction level parallelism whichcan be exploited by both super-scalar and VLIW architectures. The key to a successful general speculation strategy is a repair mechanism to handle mi...
展开
Speculative execution is a method to increase instruction level parallelism whichcan be exploited by both super-scalar and VLIW architectures. The key to a successful general speculation strategy is a repair mechanism to handle mispredicted branches and accurate reporting of exceptions for speculated instructions. Multiple instruction rollback is a technique developed for recovery from transient processor failure. Many of the difficulties encountered during recovery from branch misprediction or from instruction re-execution due to exception in a speculative execution architecture are similar to those encountered during multiple instruction rollback. The applicability of a recently developed compiler-assisted multiple instruction rollback scheme to aid in speculative execution repair is investigated. Extensions to the compiler-assisted scheme to support branch and exception repair are presented along with performance measurements across ten application programs.
收起
摘要 :
NASA's Goddard Mission Services Evolution Center (GMSEC) ground system architecture has been in development since late 2001, has successfully supported eight orbiting satellites and is being applied to many of NASA's future missio...
展开
NASA's Goddard Mission Services Evolution Center (GMSEC) ground system architecture has been in development since late 2001, has successfully supported eight orbiting satellites and is being applied to many of NASA's future missions. GMSEC can be considered an event-driven service-oriented architecture built around a publish/subscribe message bus middleware. This paper briefly discusses the GMSEC technical approaches which have led to significant cost savings and risk reduction for NASA missions operated at the Goddard Space Flight Center (GSFC). The paper then focuses on the development and operational impacts of extending the architecture across multiple mission operations facilities.
收起
摘要 :
NASA's Goddard Mission Services Evolution Center (GMSEC) ground system architecture has been in development since late 2001, has successfully supported eight orbiting satellites and is being applied to many of NASA's future missio...
展开
NASA's Goddard Mission Services Evolution Center (GMSEC) ground system architecture has been in development since late 2001, has successfully supported eight orbiting satellites and is being applied to many of NASA's future missions. GMSEC can be considered an event-driven service-oriented architecture built around a publish/subscribe message bus middleware. This paper briefly discusses the GMSEC technical approaches which have led to significant cost savings and risk reduction for NASA missions operated at the Goddard Space Flight Center (GSFC). The paper then focuses on the development and operational impacts of extending the architecture across multiple mission operations facilities.
收起
摘要 :
This viewgraph presentation gives an overview of a debugger for computational grid applications. Details are given on NAS parallel tools groups (including parallelization support tools, evaluation of various parallelization strate...
展开
This viewgraph presentation gives an overview of a debugger for computational grid applications. Details are given on NAS parallel tools groups (including parallelization support tools, evaluation of various parallelization strategies, and distributed and aggregated computing), debugger dependencies, scalability, initial implementation, the process grid, and information on Globus.
收起
摘要 :
This is the report of a workshop convened at the National Bureau of Standards on August 22-23, 1977,to discuss the different technologies applicable to computer networks serving a limited geographic area,such as a single campus,fa...
展开
This is the report of a workshop convened at the National Bureau of Standards on August 22-23, 1977,to discuss the different technologies applicable to computer networks serving a limited geographic area,such as a single campus,factory or office complex. A number of short presentations were made by active researchers and implementers in this area,afterwards the group broke up into a number of working sessions for intensive discussion of specific topics. A recorder at each session prepared a session report with the session chairman. The sessions were as follows: (1) Subnet architecture; (2) Protocols for local area networks; (3) Local network applications; (4) Network architecture; (5) Network operating systems; (6) Analysis and performance evaluation. A list of attendees and bibliography on local area computer networks is included in the report.
收起