Pub Date : 2004-04-26DOI: 10.1109/IPDPS.2004.1303273
P. Bulić, V. Gustin
There are a number of data dependence tests that have been proposed in the literature. In each test there is a different trade-off between accuracy and efficiency. The most widely used approximate data dependence tests are the Banerjee inequality and the GCD test; whereas the Omega test is a well-known exact data dependence test. We consider parallelization for microprocessors with a multimedia extension (the short SIMD execution model). For the short SIMD parallelism extraction it is essential that, if dependency exists, then the dependence distance is greater than or equal to the number of data processed in the SIMD register. This implies that some loops that could not be vectorized on traditional vector processors can still be parallelized for the short SIMD execution. In all of these tests the parallelization would be prohibited when actually there is no parallelism restriction relating to the short SIMD execution model. We present a new, fast data dependence test for array references with linear subscripts, which is used in a vectorizing compiler for microprocessors with a multimedia extension. Our test is suitable for use in a dependence analyser that is organized as a series of tests, progressively increasing in accuracy, as a replacement for the GCD or Banerjee tests.
{"title":"Fast dependence analysis in a multimedia vectorizing compiler","authors":"P. Bulić, V. Gustin","doi":"10.1109/IPDPS.2004.1303273","DOIUrl":"https://doi.org/10.1109/IPDPS.2004.1303273","url":null,"abstract":"There are a number of data dependence tests that have been proposed in the literature. In each test there is a different trade-off between accuracy and efficiency. The most widely used approximate data dependence tests are the Banerjee inequality and the GCD test; whereas the Omega test is a well-known exact data dependence test. We consider parallelization for microprocessors with a multimedia extension (the short SIMD execution model). For the short SIMD parallelism extraction it is essential that, if dependency exists, then the dependence distance is greater than or equal to the number of data processed in the SIMD register. This implies that some loops that could not be vectorized on traditional vector processors can still be parallelized for the short SIMD execution. In all of these tests the parallelization would be prohibited when actually there is no parallelism restriction relating to the short SIMD execution model. We present a new, fast data dependence test for array references with linear subscripts, which is used in a vectorizing compiler for microprocessors with a multimedia extension. Our test is suitable for use in a dependence analyser that is organized as a series of tests, progressively increasing in accuracy, as a replacement for the GCD or Banerjee tests.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125073222","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271448
Juan I. Asensio-Pérez, Y. Dimitriadis, Marta Heredia, A. Martínez-Monés, F. Álvarez, M. T. Blasco, C. Osuna
The creation of a framework of software components and their associated software design patterns would provide great benefits for the development of reusable, flexible, and customizable component-based CSCL applications. The development of such a framework implies that software developers have a proper understanding of the key concepts and principles of the domain of interest. The achievement of this understanding is particularly difficult in the CSCL domain, where there is a big separation among abstractions used by educational science experts and those used by software developers. In order to alleviate this problem, we propose, justifies, and illustrates the use of the so-called collaborative learning patterns: detailed descriptions of well-accepted types of collaborative learning activities defined by collaborative learning experts. We also present the initial steps that would be followed so that software developers identify software components applicable to several types of component-based CSCL applications. All this proposal is illustrated with the jigsaw and pyramid collaborative learning patterns and their use in the development of a real CSCL application according to the unified process software development methodology.
{"title":"Collaborative learning patterns: assisting the development of component-based CSCL applications","authors":"Juan I. Asensio-Pérez, Y. Dimitriadis, Marta Heredia, A. Martínez-Monés, F. Álvarez, M. T. Blasco, C. Osuna","doi":"10.1109/EMPDP.2004.1271448","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271448","url":null,"abstract":"The creation of a framework of software components and their associated software design patterns would provide great benefits for the development of reusable, flexible, and customizable component-based CSCL applications. The development of such a framework implies that software developers have a proper understanding of the key concepts and principles of the domain of interest. The achievement of this understanding is particularly difficult in the CSCL domain, where there is a big separation among abstractions used by educational science experts and those used by software developers. In order to alleviate this problem, we propose, justifies, and illustrates the use of the so-called collaborative learning patterns: detailed descriptions of well-accepted types of collaborative learning activities defined by collaborative learning experts. We also present the initial steps that would be followed so that software developers identify software components applicable to several types of component-based CSCL applications. All this proposal is illustrated with the jigsaw and pyramid collaborative learning patterns and their use in the development of a real CSCL application according to the unified process software development methodology.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116931603","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271435
J. Colmenar, O. Garnica, S. López, J. Hidalgo, J. Lanchares, R. Hermida
The paper has two aims: on one hand, to characterize the nature of the relationship between the latency of an asynchronous pipeline and the stage latencies when the latency of the stages is data-dependent. On the other hand, to find the closed-form expression which relates the mean value of the latency of the pipeline with the parameters which characterize the behaviour of the constituting stages. To attain these two goals, we have followed an empirical approach; we have developed a model of an asynchronous pipeline with n stages, the latency of the stages has been modelled using a probability density function and the pipeline behaviour has been simulated. From the results, we have defined linear equations which estimate the mean latency of the pipeline without needs of simulation. Finally, we have designed and implemented a 32/spl times/32 bit asynchronous pipelined multiplier, and we have tested the estimations on it.
{"title":"Empirical characterization of the latency of long asynchronous pipelines with data-dependent module delays","authors":"J. Colmenar, O. Garnica, S. López, J. Hidalgo, J. Lanchares, R. Hermida","doi":"10.1109/EMPDP.2004.1271435","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271435","url":null,"abstract":"The paper has two aims: on one hand, to characterize the nature of the relationship between the latency of an asynchronous pipeline and the stage latencies when the latency of the stages is data-dependent. On the other hand, to find the closed-form expression which relates the mean value of the latency of the pipeline with the parameters which characterize the behaviour of the constituting stages. To attain these two goals, we have followed an empirical approach; we have developed a model of an asynchronous pipeline with n stages, the latency of the stages has been modelled using a probability density function and the pipeline behaviour has been simulated. From the results, we have defined linear equations which estimate the mean latency of the pipeline without needs of simulation. Finally, we have designed and implemented a 32/spl times/32 bit asynchronous pipelined multiplier, and we have tested the estimations on it.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114057032","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271424
Bora I. Kumova
The technologies that enable and facilitate distributed simulation seam to have reached their maturity. If, however, the most significant concepts in distributed simulation, namely interest management, synchronisation and load balancing are jointly evaluated by the kernel, then the kernel can perform even better and thus the simulation time can improve further. In this contribution we present the first results of the design of a distributed simulation kernel that combines these aspects and that will be able to dynamically adapt at runtime to the simulation behaviour of any user model.
{"title":"Software design concepts of a distributed simulation kernel","authors":"Bora I. Kumova","doi":"10.1109/EMPDP.2004.1271424","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271424","url":null,"abstract":"The technologies that enable and facilitate distributed simulation seam to have reached their maturity. If, however, the most significant concepts in distributed simulation, namely interest management, synchronisation and load balancing are jointly evaluated by the kernel, then the kernel can perform even better and thus the simulation time can improve further. In this contribution we present the first results of the design of a distributed simulation kernel that combines these aspects and that will be able to dynamically adapt at runtime to the simulation behaviour of any user model.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121842990","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271432
R. Baños, C. Gil, J. Ortega, F. G. Montoya
The graph partitioning problem consists of dividing the vertices of a graph into a set of balanced parts, such that the number of edges connecting vertices in different parts is minimised. The multilevel approaches reduce the size of the graph by successively matching vertices and edges until a small graph is built, which is divided into several parts. Then simultaneous optimisation of the partitions is carried out. The complexity of the scientific applications where the graph partitioning problem appears, makes parallel processing very useful. We present a new parallel multilevel algorithm for graph partitioning, which is focused to explore different areas of the search space. This algorithm mixes heuristic techniques such as simulated annealing (SA), Tabu search (TS) and elitist mechanisms of selection. The partitions obtained by our algorithm often improve the results of the corresponding serial version, and these provided by other previously proposed algorithms.
{"title":"Parallel heuristic search in multilevel graph partitioning","authors":"R. Baños, C. Gil, J. Ortega, F. G. Montoya","doi":"10.1109/EMPDP.2004.1271432","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271432","url":null,"abstract":"The graph partitioning problem consists of dividing the vertices of a graph into a set of balanced parts, such that the number of edges connecting vertices in different parts is minimised. The multilevel approaches reduce the size of the graph by successively matching vertices and edges until a small graph is built, which is divided into several parts. Then simultaneous optimisation of the partitions is carried out. The complexity of the scientific applications where the graph partitioning problem appears, makes parallel processing very useful. We present a new parallel multilevel algorithm for graph partitioning, which is focused to explore different areas of the search space. This algorithm mixes heuristic techniques such as simulated annealing (SA), Tabu search (TS) and elitist mechanisms of selection. The partitions obtained by our algorithm often improve the results of the corresponding serial version, and these provided by other previously proposed algorithms.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133718411","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271469
E. Macías, Álvaro Suárez Sarmiento, V. Sunderam
In the last few years the use of wireless technology has increased by leaps and bounds and as a result powerful portable computers with wireless cards are viable nodes in parallel distributed computing. In this scenario it is natural to consider the possibility of frequent failures in the wireless channel. In MPI programs, such wireless network behavior is reflected as communication failure. Although the MPI standard does not handle failures, there are some projects that address this issue. To the best of our knowledge there is no previous work that presents a practical solution for fault-handling in MPI programs that run on wireless environments. We present a mechanism at the application level, that combined with wireless network monitoring software detects these failures and warns MPI applications to enable them to take appropriate action.
{"title":"Efficient monitoring to detect wireless channel failures for MPI programs","authors":"E. Macías, Álvaro Suárez Sarmiento, V. Sunderam","doi":"10.1109/EMPDP.2004.1271469","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271469","url":null,"abstract":"In the last few years the use of wireless technology has increased by leaps and bounds and as a result powerful portable computers with wireless cards are viable nodes in parallel distributed computing. In this scenario it is natural to consider the possibility of frequent failures in the wireless channel. In MPI programs, such wireless network behavior is reflected as communication failure. Although the MPI standard does not handle failures, there are some projects that address this issue. To the best of our knowledge there is no previous work that presents a practical solution for fault-handling in MPI programs that run on wireless environments. We present a mechanism at the application level, that combined with wireless network monitoring software detects these failures and warns MPI applications to enable them to take appropriate action.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115031069","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271422
D. Avresky, Y. Varoglu, M. Marinov
New network components joining running system area networks can require communication path reconfigurations. We present a novel efficient algorithm to dynamically reconfigure the network with an arbitrary topology, for the case of newly joining node. We define a restoration graph which overcomes the restriction of the up/down algorithm. We obtain an upper bound in terms of the number of messages generated in order to reconfigure the network. The termination and the liveness properties of the proposed algorithm are proven. The validation and performance improvement of the system area network (SAN) with the proposed algorithm in comparison to the up/down algorithm is presented.
{"title":"Dynamically scaling system area networks","authors":"D. Avresky, Y. Varoglu, M. Marinov","doi":"10.1109/EMPDP.2004.1271422","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271422","url":null,"abstract":"New network components joining running system area networks can require communication path reconfigurations. We present a novel efficient algorithm to dynamically reconfigure the network with an arbitrary topology, for the case of newly joining node. We define a restoration graph which overcomes the restriction of the up/down algorithm. We obtain an upper bound in terms of the number of messages generated in order to reconfigure the network. The termination and the liveness properties of the proposed algorithm are proven. The validation and performance improvement of the system area network (SAN) with the proposed algorithm in comparison to the up/down algorithm is presented.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116132742","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271428
Agathocles Gourgoulis, G. Terstyánszky, P. Kacsuk, S. Winter
We describe the implementation of a transport simulation in a parallel environment. The implementation is based on a graphical parallel programming environment called P-GRADE. The transport simulator, called MadCity, simulates a specific road network of a city and shows cars moving on the roads. To achieve scalability of the traffic simulation, the use of templates is necessary. This helps to control the number of participating processes required for the simulation without making modifications to the simulator's source code. Performance results are collected from four, eight and sixteen nodes of the Parsifal cluster and compared with the sequential execution results of the simulator. The implementation of the transport simulator is extended further to support the simulation of multiple cities within the same cluster and on the Grid.
{"title":"Creating scalable traffic simulation on clusters","authors":"Agathocles Gourgoulis, G. Terstyánszky, P. Kacsuk, S. Winter","doi":"10.1109/EMPDP.2004.1271428","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271428","url":null,"abstract":"We describe the implementation of a transport simulation in a parallel environment. The implementation is based on a graphical parallel programming environment called P-GRADE. The transport simulator, called MadCity, simulates a specific road network of a city and shows cars moving on the roads. To achieve scalability of the traffic simulation, the use of templates is necessary. This helps to control the number of participating processes required for the simulation without making modifications to the simulator's source code. Performance results are collected from four, eight and sixteen nodes of the Parsifal cluster and compared with the sequential execution results of the simulator. The implementation of the transport simulator is extended further to support the simulation of multiple cities within the same cluster and on the Grid.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133704022","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271450
A. Riera, M. Lama, E. S. Vila, R. Amorim, X. Sobrino, S. Barro
We present a multi-agent system (MAS) communication protocol for an intelligent learning management system that provides, among others, collaborative services. The motivation behind this study stems from the need to enhance the possibilities of the current software architecture of the EUME project. Our aim is to improve intelligent resource management and communication efficiency. Our approach consists of applying the FIPA agent paradigm to the entire architecture, thus coming up with a fully multi-agent system. To show how the new system works, we also provide a study of communications in a number of typical collaborative scenarios.
{"title":"Study of communication in a multi-agent system for collaborative learning scenarios","authors":"A. Riera, M. Lama, E. S. Vila, R. Amorim, X. Sobrino, S. Barro","doi":"10.1109/EMPDP.2004.1271450","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271450","url":null,"abstract":"We present a multi-agent system (MAS) communication protocol for an intelligent learning management system that provides, among others, collaborative services. The motivation behind this study stems from the need to enhance the possibilities of the current software architecture of the EUME project. Our aim is to improve intelligent resource management and communication efficiency. Our approach consists of applying the FIPA agent paradigm to the entire architecture, thus coming up with a fully multi-agent system. To show how the new system works, we also provide a study of communications in a number of typical collaborative scenarios.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115472737","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}
Pub Date : 2004-03-08DOI: 10.1109/EMPDP.2004.1271461
M. Díaz, B. Rubio, E. Soler, J. M. Troya
SBASCO is a new programming environment for the development of parallel and distributed high-performance scientific applications. The approach integrates both skeleton-based and component technologies. The main goal of the proposal is to provide a high-level programmability system for the efficient development of numerical applications with performance portability on different platforms. We present the system programming model which considers two different views of a component interface: one from the point of view of the application programmer and another thought to be used by a configuration tool in order to establish efficient implementations. This can be achieved due to the knowledge at the interface level of data distribution and processor layout inside each component. The programming model borrows from software skeletons a cost model enhanced by a run-time analysis, which enables one to automatically establish a suitable degree of parallelism and replication of the internal structure of a component.
{"title":"SBASCO: skeleton-based scientific components","authors":"M. Díaz, B. Rubio, E. Soler, J. M. Troya","doi":"10.1109/EMPDP.2004.1271461","DOIUrl":"https://doi.org/10.1109/EMPDP.2004.1271461","url":null,"abstract":"SBASCO is a new programming environment for the development of parallel and distributed high-performance scientific applications. The approach integrates both skeleton-based and component technologies. The main goal of the proposal is to provide a high-level programmability system for the efficient development of numerical applications with performance portability on different platforms. We present the system programming model which considers two different views of a component interface: one from the point of view of the application programmer and another thought to be used by a configuration tool in order to establish efficient implementations. This can be achieved due to the knowledge at the interface level of data distribution and processor layout inside each component. The programming model borrows from software skeletons a cost model enhanced by a run-time analysis, which enables one to automatically establish a suitable degree of parallelism and replication of the internal structure of a component.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128515159","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：