A programmable Java distributed system, which adapts to available resources, has been developed to minimise the overall processing time of computationally intensive problems. The system exploits the free resources of a heterogeneous set of computers linked together by a network, communicating using SUN Microsystems' Remote Method Invocation and Java sockets. It uses a multi-tiered distributed system model, which in principal allows for a system of unbounded size. The system consists of an n-ary tree of nodes where the internal nodes perform the scheduling and the leaves do the processing. The scheduler nodes communicate in a peer-to-peer manner and the processing nodes operate in a strictly client-server manner with their respective scheduler. The independent schedulers on each tier of the tree dynamically allocate resources between problems based on the constantly changing characteristics of the underlying network. The system has been evaluated over a network of 86 PCs with a bioinformatics application and the travelling salesman optimisation problem.
{"title":"Adaptive scheduling across a distributed computation platform","authors":"Andrew J. Page, Thomas M. Keane, T. Naughton","doi":"10.1109/ISPDC.2004.8","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.8","url":null,"abstract":"A programmable Java distributed system, which adapts to available resources, has been developed to minimise the overall processing time of computationally intensive problems. The system exploits the free resources of a heterogeneous set of computers linked together by a network, communicating using SUN Microsystems' Remote Method Invocation and Java sockets. It uses a multi-tiered distributed system model, which in principal allows for a system of unbounded size. The system consists of an n-ary tree of nodes where the internal nodes perform the scheduling and the leaves do the processing. The scheduler nodes communicate in a peer-to-peer manner and the processing nodes operate in a strictly client-server manner with their respective scheduler. The independent schedulers on each tier of the tree dynamically allocate resources between problems based on the constantly changing characteristics of the underlying network. The system has been evaluated over a network of 86 PCs with a bioinformatics application and the travelling salesman optimisation problem.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86448528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
E. Anceaume, C. Delporte-Gallet, H. Fauconnier, M. Hurfin, G. L. Lann
In this paper, we propose the scattered byzantine failure model. In this model, processes alternate correct and faulty periods. Specifically, during its faulty periods, a process behaves arbitrarily (one cannot expect anything from it during these periods) whereas during its correct periods, it behaves according to its specification. In that sense, the scattered byzantine failure model generalizes the classical Byzantine failure model. We characterize two reliable services guaranteeing timeliness properties in the presence of Byzantine failures, namely the clock synchronization and the /spl delta/-Atomic Broadcast. We identify necessary and sufficient conditions to ensure the correctness of both services in the scattered byzantine failure model.
{"title":"Designing modular services in the scattered byzantine failure model","authors":"E. Anceaume, C. Delporte-Gallet, H. Fauconnier, M. Hurfin, G. L. Lann","doi":"10.1109/ISPDC.2004.18","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.18","url":null,"abstract":"In this paper, we propose the scattered byzantine failure model. In this model, processes alternate correct and faulty periods. Specifically, during its faulty periods, a process behaves arbitrarily (one cannot expect anything from it during these periods) whereas during its correct periods, it behaves according to its specification. In that sense, the scattered byzantine failure model generalizes the classical Byzantine failure model. We characterize two reliable services guaranteeing timeliness properties in the presence of Byzantine failures, namely the clock synchronization and the /spl delta/-Atomic Broadcast. We identify necessary and sufficient conditions to ensure the correctness of both services in the scattered byzantine failure model.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80767991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper gives an overview of the material to be discussed in the invited keynote presentation by H. J. Siegel. Performing computing and communication tasks on parallel and distributed systems involves the coordinated use of different types of machines, networks, interfaces, and other resources. Decisions about how best to allocate resources are often based on estimated values of task and system parameters, due to uncertainties in the system environment. An important research problem is the development of resource management strategies that can guarantee a particular system performance given such uncertainties. We have designed a methodology for deriving the degree of robustness of a resource allocation - the maximum amount of collective uncertainty in system parameters within which a user-specified level of system performance (QoS) can be guaranteed. Our four-step procedure for deriving a robustness metric for an arbitrary system will be presented. We will illustrate this procedure and its usefulness by deriving robustness metrics for some example distributed systems.
{"title":"The robustness of resource allocation in parallel and distributed computing systems","authors":"Shoukat Ali, H. Siegel, A. A. Maciejewski","doi":"10.1109/ISPDC.2004.51","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.51","url":null,"abstract":"This paper gives an overview of the material to be discussed in the invited keynote presentation by H. J. Siegel. Performing computing and communication tasks on parallel and distributed systems involves the coordinated use of different types of machines, networks, interfaces, and other resources. Decisions about how best to allocate resources are often based on estimated values of task and system parameters, due to uncertainties in the system environment. An important research problem is the development of resource management strategies that can guarantee a particular system performance given such uncertainties. We have designed a methodology for deriving the degree of robustness of a resource allocation - the maximum amount of collective uncertainty in system parameters within which a user-specified level of system performance (QoS) can be guaranteed. Our four-step procedure for deriving a robustness metric for an arbitrary system will be presented. We will illustrate this procedure and its usefulness by deriving robustness metrics for some example distributed systems.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"78200642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
G. Agosta, S. Crespi-Reghizzi, Gerlando Falauto, M. Sykora
The application fields of bytecode virtual machines and VLIW processors overlap in the area of embedded and mobile systems, where the two technologies offer different benefits, namely high code portability, low power consumption and reduced hardware cost. Dynamic compilation makes it possible to bridge the gap between the two technologies, but special attention must be paid to software instruction scheduling, a must for the VLIW architectures. We have implemented JIST, a Virtual Machine and JIT compiler for Java Bytecode targeted to a VLIW processor. We show the impact of various optimizations on the performance of code compiled with JIST through the experimental study on a set of benchmark programs. We report significant speedups, and increments in the number of instructions issued per cycle up to 50% with respect to the non-scheduling version of the JIT compiler. Further optimizations are discussed.
{"title":"JIST: just-in-time scheduling translation for parallel processors","authors":"G. Agosta, S. Crespi-Reghizzi, Gerlando Falauto, M. Sykora","doi":"10.1109/ISPDC.2004.32","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.32","url":null,"abstract":"The application fields of bytecode virtual machines and VLIW processors overlap in the area of embedded and mobile systems, where the two technologies offer different benefits, namely high code portability, low power consumption and reduced hardware cost. Dynamic compilation makes it possible to bridge the gap between the two technologies, but special attention must be paid to software instruction scheduling, a must for the VLIW architectures. We have implemented JIST, a Virtual Machine and JIT compiler for Java Bytecode targeted to a VLIW processor. We show the impact of various optimizations on the performance of code compiled with JIST through the experimental study on a set of benchmark programs. We report significant speedups, and increments in the number of instructions issued per cycle up to 50% with respect to the non-scheduling version of the JIT compiler. Further optimizations are discussed.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91502448","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Mobile devices, such as mobile phones and PDAs, have gained wide-spread popularity. Applications for this kind of mobile devices have to adapt to changes in context, such as variations in network bandwidth, battery power, connectivity, reachability of services and hosts, and so on. In this paper, we define context-aware action systems that provides a systematic method for managing and processing context information. The meaning of context-aware action systems is defined in terms of classical action systems, so that the properties of context-aware action systems can be proved using standard action systems proof techniques. We describe the essential notions of this formalism and illustrate the framework with examples on context-aware services for mobile applications.
{"title":"A formalism for context-aware mobile computing","authors":"Lu Yan, K. Sere","doi":"10.1109/ISPDC.2004.1","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.1","url":null,"abstract":"Mobile devices, such as mobile phones and PDAs, have gained wide-spread popularity. Applications for this kind of mobile devices have to adapt to changes in context, such as variations in network bandwidth, battery power, connectivity, reachability of services and hosts, and so on. In this paper, we define context-aware action systems that provides a systematic method for managing and processing context information. The meaning of context-aware action systems is defined in terms of classical action systems, so that the properties of context-aware action systems can be proved using standard action systems proof techniques. We describe the essential notions of this formalism and illustrate the framework with examples on context-aware services for mobile applications.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89091246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes the work of an objective framework designed to be used in the parallelization of a set of related algorithms. As a concrete application a parallel ant colony optimization algorithm (ACO) for the travelling salesman problem (TSP) is presented. The idea behind the system we are describing is to have a reusable framework for running several sequential algorithms in a parallel environment. The algorithms that the framework can be used with have several things in common: they have to run in cycles and the work should be possible to be split between several "processing units". The parallel framework uses the message-passing communication paradigm and is organized as a master-slave system. The ACO for TSP implemented by means of the parallel framework proves to have good performances: approximately linear speedup and low communication cost.
{"title":"Parallel framework for ant-like algorithms","authors":"M. Craus, Laurentiu Rudeanu","doi":"10.1109/ISPDC.2004.37","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.37","url":null,"abstract":"This paper describes the work of an objective framework designed to be used in the parallelization of a set of related algorithms. As a concrete application a parallel ant colony optimization algorithm (ACO) for the travelling salesman problem (TSP) is presented. The idea behind the system we are describing is to have a reusable framework for running several sequential algorithms in a parallel environment. The algorithms that the framework can be used with have several things in common: they have to run in cycles and the work should be possible to be split between several \"processing units\". The parallel framework uses the message-passing communication paradigm and is organized as a master-slave system. The ACO for TSP implemented by means of the parallel framework proves to have good performances: approximately linear speedup and low communication cost.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84771830","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
L. A. Bongo, Otto J. Anshus, J. Bjørndalen, Tore Larsen
We identify two ways of increasing the performance of allreduce-style of collective operations in a multi-cluster with large WAN latencies: (i) hiding latency in system noise, and (ii) conditional-allreduce where knowledge about the application is used to reduce the number of WAN messages. In our multicluster, system noise was not large enough to hide the WAN latency. But, the latency could be hidden using conditional-allreduce, since on many iterations only cluster-local values were needed, and many of the values needed from other clusters were prefetched. A speedup of 2.4 was achieved for a microbenchmark. Prefetching introduced a small overhead in the cluster with the slowest hosts.
{"title":"Extending collective operations with application semantics for improving multi-cluster performance","authors":"L. A. Bongo, Otto J. Anshus, J. Bjørndalen, Tore Larsen","doi":"10.1109/ISPDC.2004.24","DOIUrl":"https://doi.org/10.1109/ISPDC.2004.24","url":null,"abstract":"We identify two ways of increasing the performance of allreduce-style of collective operations in a multi-cluster with large WAN latencies: (i) hiding latency in system noise, and (ii) conditional-allreduce where knowledge about the application is used to reduce the number of WAN messages. In our multicluster, system noise was not large enough to hide the WAN latency. But, the latency could be hidden using conditional-allreduce, since on many iterations only cluster-local values were needed, and many of the values needed from other clusters were prefetched. A speedup of 2.4 was achieved for a microbenchmark. Prefetching introduced a small overhead in the cluster with the slowest hosts.","PeriodicalId":62714,"journal":{"name":"骈文研究","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-07-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90783786","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：