作者:
Benz, HDepartment of Computer Science
Institute of Parallel and Distributed High Performance Systems University of Stuttgart Breitwiesenstr. 20-22 Stuttgart D-70565 Germany
This paper presents a methodology for performance prediction of parallel algorithms and illustrates its use on a large scale computational chemistry application. The performance prediction uses a component time charac...
详细信息
This paper presents a methodology for performance prediction of parallel algorithms and illustrates its use on a large scale computational chemistry application. The performance prediction uses a component time characterization technique which splits up the sequential code into computational components and measures the time for each of them. The parallel algorithm is built from these components by adding communication routines. A “Processor Activity Graph” (PAG) providing a graphical representation of the parallel algorithm runtime behaviour is used for predicting the execution time. For a case study a Self Consistent Field (SCF) computation has been selected which forms the basis of many computational chemistry packages [4, 5]. The performance model of SCF computation has been built and the prediction have been compared with the results of measurements. The measurements have been provided on a mesh connected distributed memory parallelcomputer (128 T800 Parsytec SuperCluster). The prediction error is less than 10%. Performance optimisation of the application has been achieved by reducing the communication overhead and changing the data representation.
A Health Care Establishment (HCE) is an establishment where medical services are rendered. The above services are provided by the health care personnel. The infrastructure of a HCE may include Information Technology (...
详细信息
Object-oriented development is a structuring paradigm. It does not provide substantial support for the specification of ‘end to end’ control and data flow. This has implications for hard real-time system development...
详细信息
Traditional configurable operating systems typically provide a fixed and limited set of functionality. We propose a metalevel architecture, where application-defined objects can choose from a rich selection of possibl...
详细信息
Traditional configurable operating systems typically provide a fixed and limited set of functionality. We propose a metalevel architecture, where application-defined objects can choose from a rich selection of possible configurations and are therefore able to dynamically change the way in which they are executed to the configuration that suits best. This allows applications to adapt operating system behaviour to even unanticipated requirements during run-time. A distributed application which processes multimedia data serves as an example to illustrate the concepts described.
The paper presents a strategy for the composition of components to meet 'end to end' timing requirements. This is a framework based approach known as transaction integration. A framework is constructed through...
详细信息
The paper presents a strategy for the composition of components to meet 'end to end' timing requirements. This is a framework based approach known as transaction integration. A framework is constructed through the use of transactions. A transaction represents a control and data flow path across a set of component interfaces, or placeholders, that must be completed by a specific deadline. Reusable real time components are 'plugged' into each placeholder. The paper provides techniques to verify that the performance characteristics of each placeholder, and the 'end to end' deadlines associated with each transaction are met.
Genericity of parallel programming environments, enabling development of portable parallel programs, is expected to result in performance penalties. Furthermore, programmability and tool support of programming environ...
Genericity of parallel programming environments, enabling development of portable parallel programs, is expected to result in performance penalties. Furthermore, programmability and tool support of programming environments are important issues if a choice between programming environments has to be made. In this paper we propose a methodology to compare native and generic parallel programming environments, taking into account such competing issues as portability and performance. As a case study, this paper compares the Iserver-Occam, Parix, Express and PVM parallel programming environments on a 512-node Parasytec GCel. Furthermore, we apply our methodology to compare Parix and PVM on a new architecture, a 32-node Parsytec PowerXplorer, which is based on the PowerPC chip. In our approach we start with a representative application and isolate the basic (environment)-dependent building blocks. These basic building blocks, which depend on floating-point performance and communication capabilities of the environments, are analysed independently. We have measured point-to-point communication times, global communication times and floating-point performance. All information is combined into a time complexity analysis, allowing comparison of the environments on different degrees of functionality. Together with demands for portability of the code and development time (i.e. programmability), an overall judgement of the environments is given.
Traditional object-oriented real-time systems are often limited an that they provide only one approach to real-time object support. Taking the increasing demand for flexible and extensible object support environments ...
详细信息
Traditional object-oriented real-time systems are often limited an that they provide only one approach to real-time object support. Taking the increasing demand for flexible and extensible object support environments ...
详细信息
Traditional object-oriented real-time systems are often limited an that they provide only one approach to real-time object support. Taking the increasing demand for flexible and extensible object support environments into account, we discuss the design and implementation of a small object-oriented real-time executive based on a sub-framework which we call Roo. Roo is a component of the Tigger framework (our proposal for an extensible object support operating system) and is intended to support different object models providing soft real-time behaviour. Roo provides support for different mechanisms and policies for real-time thread management, scheduling and synchronization. In this it serves as a basis for other components of the Tigger framework.< >
A distributed environment for cooperative software (DECS) provides a flexible and extensible platform for the development and execution of cooperative software. It takes the advantages of object-oriented technology an...
详细信息
A distributed environment for cooperative software (DECS) provides a flexible and extensible platform for the development and execution of cooperative software. It takes the advantages of object-oriented technology and emphasizes composition as the basis for developing large distributed system from reusable components. In this paper, we give an overview of DECS, present the compositional approach, and then describe the design and implementation of DECS run-time system.
暂无评论