Taking advantage of the Grid for high-performance scientific computing requires a software framework layered above the Grid that supports a new style of programming, component-based programming. the component architec...
详细信息
Taking advantage of the Grid for high-performance scientific computing requires a software framework layered above the Grid that supports a new style of programming, component-based programming. the component architecture toolkit (CAT) facilitates such programming by providing a conceptually simple `port-based' component model together with a suite of developer tools for incorporating existing scientific codes into the framework, as well as a set of end user tools for locating, composing, building and running distributed component application over the Grid.
In this study, the impact of a variety of features provided by emerging system-area networks on the implementation of Software Distributed Shared Memory (SDSM), and on the Cashmere system in particular, is evaluated. ...
详细信息
In this study, the impact of a variety of features provided by emerging system-area networks on the implementation of Software Distributed Shared Memory (SDSM), and on the Cashmere system in particular, is evaluated. the results suggest that for systems of modest size, low latency is much more important for SDSM performancethan are remote writes, broadcast, or total ordering. In addition, results on an emulated 32-node system indicate that broadcast based on remote writes of widely-shared data may improve performance by up to 51% for some applications.
Seamless access to different systems of different vendors at different sites is an important prerequisite to effective and efficient use of distributed resources. Learning about new systems, new software, and new inte...
详细信息
Seamless access to different systems of different vendors at different sites is an important prerequisite to effective and efficient use of distributed resources. Learning about new systems, new software, and new interfaces is a time-consuming task for users who actually want to run their applications. UNICORE is a project to overcome these difficulties by providing a uniform interface for job preparation and control which gives seamless and secure access to supercomputer resources. It is an ambitious project delivering a production ready prototype within two years. the presentation will focus on the UNICORE architecture, especially the protocol and the underlying security mechanisms.
the problem of low-contrast radiologic image analysis is connected with visual identification impossibility of 1% (or less) in brightness variations, whereas for tasks of medical diagnosis, much higher sensitivity is ...
详细信息
the problem of low-contrast radiologic image analysis is connected with visual identification impossibility of 1% (or less) in brightness variations, whereas for tasks of medical diagnosis, much higher sensitivity is needed. To solve this problem, pseudocoherent holographic interferometry method (PHIM) is developed. the idea of PHIM is related to the presentation of a low-contrast image as a topologic surface with further illumination by means of a virtual pseudocoherent optical wave. Interaction of the reflected subject wave withthe support one forms a new complex interferometric image. the important differences of PHIM from optical holographic interferometry are discussed.
Earlier research has shown that adaptive routing can help in improving network performance. However it has not received adequate attention in commercial routers mainly due to the additional hardware complexity and the...
详细信息
ISBN:
(纸本)0769500048
Earlier research has shown that adaptive routing can help in improving network performance. However it has not received adequate attention in commercial routers mainly due to the additional hardware complexity and the perceived cost and performance degradation that may result from this complexity: these concerns can be mitigated if one can design a cost-effective router that can support adaptive routing. this paper proposes a three step recipe - Look-Ahead routing, intelligent Path Selection, and an Economic Storage implementation, called the LAPSES approach - for cost-effective highperformance pipelined adaptive router design. the first step, look-ahead routing, reduces a pipeline stage in the router by making table lookup and arbitration concurrent. Next, three new traffic-sensitive path selection heuristics (LRU, LFU and MAX-CREDIT) are proposed to select one of the available alternate paths. Finally, two techniques for reducing routing table size of the adaptive router are presented. these are called meta-table routing and economical storage. the proposed economical storage needs a routing table with only 9 and 27 entries for two and three dimensional meshes, respectively! All these design ideas are evaluated on a (16 x 16) mesh network via simulation. A fully adaptive algorithm and various traffic patterns are used to examine the performance benefits. performance results show that the look-ahead design as well as the path selection heuristics boost network performance, while the economical storage approach turns out to be an ideal choice in comparison to full-table and meta-table options. We believe the router resulting from these three design enhancements can make adaptive routing a viable choice for interconnects.
this paper describes work in progress to develop a standard for interoperability among high-performance scientific components. this research stems from growing recognition that the scientific community needs to better...
详细信息
this paper describes work in progress to develop a standard for interoperability among high-performance scientific components. this research stems from growing recognition that the scientific community needs to better manage the complexity of multidisciplinary simulations and better address scalable performance issues on parallel and distributed architectures. Driving forces are the need for fast connections among components that perform numerically intensive work and for parallel collective interactions among components that use multiple processes or threads. this paper focuses on the areas we believe are most crucial in this context, namely, an interface definition language that supports scientific abstractions for specifying component interfaces and a ports connection model for specifying component interactions.
RSD (Resource and Service Description) is a software architecture for specifying, registering and accessing resources and services in complex heterogeneous computing environments. It has three major components: a grap...
详细信息
RSD (Resource and Service Description) is a software architecture for specifying, registering and accessing resources and services in complex heterogeneous computing environments. It has three major components: a graphical resource editor, a compiler system for translating textual specifications, and an application programming interface. In this paper we focus on the language definition and the compiler system that generates portable data objects for registering and accessing distributed resources. We also present the mechanisms used by RSD to maintain information on the status of dynamic resources such as CPU load or network bandwidth. RSD is a purely descriptive tool that provides up-to-date information on system resources. Other tools, like resource brokers, mappers or job migrators, build on it.
A software package is introduced to aid in the teaching of courses in parallel processing, particularly topics such as software distributed shared memory, memory consistency models, and synchronization mechanisms. the...
详细信息
CORBA applications can transparently use service instances running on the client's machine, on the local-area network, or across the Internet. Standard CORBA services help the application locate service instances,...
详细信息
CORBA applications can transparently use service instances running on the client's machine, on the local-area network, or across the Internet. Standard CORBA services help the application locate service instances, but do not provide a mechanism to identify service instances that will give good performance. the performanceBroker executes performance test suites on application service instances and selects service instances that will give superior application performance. the Broker weights performance test results according to client-specified criteria to choose the service instances that will provide the best application performance and allocates those service instances to clients. Tests with a distributed ray tracing application show that service instances chosen by the performanceBroker give better performancethan service instances chosen by round-robin or random selection in local-area network and Internet environments.
the paper presents a computerarchitecture Learning and Knowledge Assessment System named the CALKAS. It is a software tool aimed to be used for teaching computerarchitecture and organization. It offers the knowledge...
详细信息
暂无评论