In this paper we present a hierarchy of bitstreams repositories for FPGA-based networked and partially reconfigurable systems. These systems target embedded systems with very scarce hardware resources taking advantage of dynamic, specific and optimized architectures. Based on FPGA integrated circuits, they require a single FPGA with a network controller and less external memories to store reconfiguration software, bitstreams and buffer pools used by today¿s standard communication protocols. Our measures, based on a real implementation, show that our repository hierarchy is functional and can download bitstreams with a reconfiguration speed ten times faster than known solutions.
{"title":"Bitstreams Repository Hierarchy for FPGA Partially Reconfigurable Systems","authors":"P. Bomel, J. Diguet, G. Gogniat, J. Crenne","doi":"10.1109/ISPDC.2008.14","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.14","url":null,"abstract":"In this paper we present a hierarchy of bitstreams repositories for FPGA-based networked and partially reconfigurable systems. These systems target embedded systems with very scarce hardware resources taking advantage of dynamic, specific and optimized architectures. Based on FPGA integrated circuits, they require a single FPGA with a network controller and less external memories to store reconfiguration software, bitstreams and buffer pools used by today¿s standard communication protocols. Our measures, based on a real implementation, show that our repository hierarchy is functional and can download bitstreams with a reconfiguration speed ten times faster than known solutions.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"141 2 Suppl 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124880213","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}
Self-stabilizing token circulation algorithms are not always adapted for dynamic networks. Random walks are well known to play a crucial role in the design of randomized algorithms. The combination of these two concepts makes it possible to design a solution that is adaptive to topology changes and is tolerant to transient faults. Our purpose in this paper is to study the behavior of such algorithms with possible transient failures. We provide a probabilistic analysis of the waiting time. More precisely, we give two bounds on the probability for a processor to wait for the token more than a certain amount of time. The first bound is based on the token return time (the expected time for the token to visit again a processor) and the second one, a tighter upper bound, is based on its variance. Next, we characterize a local ¿criterion of suspicion¿ for each processor to be in a faulty global configuration; in fact a token loss detection. Thanks to this criterion, we propose to refine the timeout procedure used in [6, 1]. Thus, an improved version of an adaptive and self-stabilizing random walk token circulation algorithm can be designed.
{"title":"Token Loss Detection for Random Walk based Algorithm","authors":"Thibault Bernard, A. Bui, D. Sohier","doi":"10.1109/ISPDC.2008.70","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.70","url":null,"abstract":"Self-stabilizing token circulation algorithms are not always adapted for dynamic networks. Random walks are well known to play a crucial role in the design of randomized algorithms. The combination of these two concepts makes it possible to design a solution that is adaptive to topology changes and is tolerant to transient faults. Our purpose in this paper is to study the behavior of such algorithms with possible transient failures. We provide a probabilistic analysis of the waiting time. More precisely, we give two bounds on the probability for a processor to wait for the token more than a certain amount of time. The first bound is based on the token return time (the expected time for the token to visit again a processor) and the second one, a tighter upper bound, is based on its variance. Next, we characterize a local ¿criterion of suspicion¿ for each processor to be in a faulty global configuration; in fact a token loss detection. Thanks to this criterion, we propose to refine the timeout procedure used in [6, 1]. Thus, an improved version of an adaptive and self-stabilizing random walk token circulation algorithm can be designed.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128116961","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}
The increasing power of computation of modern processors rely on the increasing number of cores per chip. The challenge of software developers is to keep this power with the legacy code. Although commercial and non commercial libraries are improving their codes step by step, there exits probably insurmountable scalability issues for standard programming models due to the fact that using locks to implement synchronisation is inherently a bottleneck. We propose an implementation of the divide and conquer algorithm to compute the eigenpairs of symmetric tridiagonal matrices on multicore systems. We take advantage of the natural parallelism of the method by using pthreads. We avoided as much as possible the negative impact of synchronisation in the performance by overlapping operations of different classes. Furthermore, the unevenly workload distribution of the computational cost of the elemental tasks yields in a speedup even larger than expected.
{"title":"A Threaded Divide and Conquer Symmetric Tridiagonal Eigensolver on Multicore Systems","authors":"A. Vidal, M. Boratto, P. Alonso","doi":"10.1109/ISPDC.2008.44","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.44","url":null,"abstract":"The increasing power of computation of modern processors rely on the increasing number of cores per chip. The challenge of software developers is to keep this power with the legacy code. Although commercial and non commercial libraries are improving their codes step by step, there exits probably insurmountable scalability issues for standard programming models due to the fact that using locks to implement synchronisation is inherently a bottleneck. We propose an implementation of the divide and conquer algorithm to compute the eigenpairs of symmetric tridiagonal matrices on multicore systems. We take advantage of the natural parallelism of the method by using pthreads. We avoided as much as possible the negative impact of synchronisation in the performance by overlapping operations of different classes. Furthermore, the unevenly workload distribution of the computational cost of the elemental tasks yields in a speedup even larger than expected.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116548677","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 discusses the design and the implementation of the LU factorization routines included in the Heterogeneous ScaLAPACK library, which is built on top of ScaLAPACK. These routines are used in the factorization and solution of a dense system of linear equations. They are implemented using optimized PBLAS, BLACS and BLAS libraries for heterogeneous computational clusters. We present the details of the implementation as well as performance results on a heterogeneous computing cluster.
{"title":"Scalable Dense Factorizations for Heterogeneous Computational Clusters","authors":"Ravi Reddy, Alexey L. Lastovetsky, P. Alonso","doi":"10.1109/ISPDC.2008.10","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.10","url":null,"abstract":"This paper discusses the design and the implementation of the LU factorization routines included in the Heterogeneous ScaLAPACK library, which is built on top of ScaLAPACK. These routines are used in the factorization and solution of a dense system of linear equations. They are implemented using optimized PBLAS, BLACS and BLAS libraries for heterogeneous computational clusters. We present the details of the implementation as well as performance results on a heterogeneous computing cluster.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"468 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124380574","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}
The advantages of integrating reinforcement learning (RL) techniques into scientific parallel time-stepping applications have been revealed in research work over the past few years. The object of the integration was to automatically select the most appropriate dynamic loop scheduling (DLS) algorithm from a set of available algorithms with the purpose of improving the application performance via load balancing during the application execution. This paper investigates the performance of such a dynamic loop scheduling with reinforcement learning (DLS-with-RL) approach to load balancing. The DLS-with-RL is most suitable for use in time-stepping scientific applications with large number of steps. The RL agent's characteristics depend on a learning rate parameter and a discount factor parameter. An application simulating wavepacket dynamics that incorporates a DLS-with-RL approach is allowed to execute on a cluster of workstations to investigate the influences of these parameters. The RL agent implemented two RL algorithms: QLEARN and SARSA learning. Preliminary results indicate that on a fixed number of processors, the simulation completion time is not sensitive to the values of the learning parameters used in the experiments. The results also indicate that for this application, there is no advantage of choosing one RL technique over another, even though the techniques differed significantly in the number of times they selected the various DLS algorithms.
{"title":"Investigating a Dynamic Loop Scheduling with Reinforcement Learning Approach to Load Balancing in Scientific Applications","authors":"M. Rashid, I. Banicescu, R. Cariño","doi":"10.1109/ISPDC.2008.25","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.25","url":null,"abstract":"The advantages of integrating reinforcement learning (RL) techniques into scientific parallel time-stepping applications have been revealed in research work over the past few years. The object of the integration was to automatically select the most appropriate dynamic loop scheduling (DLS) algorithm from a set of available algorithms with the purpose of improving the application performance via load balancing during the application execution. This paper investigates the performance of such a dynamic loop scheduling with reinforcement learning (DLS-with-RL) approach to load balancing. The DLS-with-RL is most suitable for use in time-stepping scientific applications with large number of steps. The RL agent's characteristics depend on a learning rate parameter and a discount factor parameter. An application simulating wavepacket dynamics that incorporates a DLS-with-RL approach is allowed to execute on a cluster of workstations to investigate the influences of these parameters. The RL agent implemented two RL algorithms: QLEARN and SARSA learning. Preliminary results indicate that on a fixed number of processors, the simulation completion time is not sensitive to the values of the learning parameters used in the experiments. The results also indicate that for this application, there is no advantage of choosing one RL technique over another, even though the techniques differed significantly in the number of times they selected the various DLS algorithms.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114701608","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}
A new parallel algorithm for finding the frequent itemsets in databases is presented. It differs fundamentally of well known Apriori algorithm, where at the beginning of every step, the dimension of the new frequent itemsets increases by 1 . In our algorithm the frequent itemsets are determined by progressively enlarging the interval which the individual items appertain, i.e. if at the k-th step the new candidates are from [i, i+k] intervals, i=1, 2,..., n-k, at the next step, k+1, the new candidates will belong to [i, i+k+1] intervals, i=1, 2,..., n-k-1. The frequent individual items are identified by their index. The basic idea is that the new frequent itemsets with individual items from the interval [i, j], simultaneously contain the items i and j. The frequent itemsets are built by sharing the work between n processors. Hereby, the processor Pi computes, step by step, the sets Fi,j of the frequent itemsets with individual items from the intervals [i, j], j=i,..., n. In order to compute the set Fi,j, the processing unit Pi uses Fi,j-1 obtained in the previous step and Fi+1,j received from the processor Pi+1. The main advantage of our parallel algorithm is that it uses a communication pattern known before algorithm start, which allows mapping communication to hardware. Another major advantage is that the set of the transactions can be distributed to processors prior to beginning. This is possible because a processor Pi has to compute Fi,j, j=i,..., n and therefore only the transactions containing the frequent item i are needed.
提出了一种新的查找数据库频繁项集的并行算法。它与众所周知的Apriori算法有根本的不同，在每一步开始时，新的频繁项集的维数增加1。在我们的算法中，频繁项集是通过逐步扩大单个项所属的区间来确定的，即如果在第k步，新的候选项来自[i, i+k]个区间，则i= 1,2，…， n-k，在下一步，k+1，新的候选将属于[i, i+k+1]区间，i= 1,2，…, n - k - 1。频繁出现的单个项目由它们的索引标识。其基本思想是新的频繁项集包含区间[i, j]中的单个项，同时包含项i和项j。频繁项集通过在n个处理器之间共享工作来构建。由此，处理器Pi逐级计算区间[i, j]， j=i，…中单个项目的频繁项集的集合Fi,j。， n。为了计算集合Fi,j，处理单元Pi使用上一步得到的Fi,j-1和处理器Pi+1接收到的Fi+1,j。我们的并行算法的主要优点是，它使用在算法开始之前就已知的通信模式，这允许将通信映射到硬件。另一个主要优点是，事务集可以在开始之前分发给处理器。这是可能的，因为处理器Pi必须计算Fi,j, j=i，…， n，因此只需要包含频繁项I的事务。
{"title":"A New Parallel Algorithm for the Frequent Itemset Mining Problem","authors":"M. Craus","doi":"10.1109/ISPDC.2008.45","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.45","url":null,"abstract":"A new parallel algorithm for finding the frequent itemsets in databases is presented. It differs fundamentally of well known Apriori algorithm, where at the beginning of every step, the dimension of the new frequent itemsets increases by 1 . In our algorithm the frequent itemsets are determined by progressively enlarging the interval which the individual items appertain, i.e. if at the k-th step the new candidates are from [i, i+k] intervals, i=1, 2,..., n-k, at the next step, k+1, the new candidates will belong to [i, i+k+1] intervals, i=1, 2,..., n-k-1. The frequent individual items are identified by their index. The basic idea is that the new frequent itemsets with individual items from the interval [i, j], simultaneously contain the items i and j. The frequent itemsets are built by sharing the work between n processors. Hereby, the processor Pi computes, step by step, the sets Fi,j of the frequent itemsets with individual items from the intervals [i, j], j=i,..., n. In order to compute the set Fi,j, the processing unit Pi uses Fi,j-1 obtained in the previous step and Fi+1,j received from the processor Pi+1. The main advantage of our parallel algorithm is that it uses a communication pattern known before algorithm start, which allows mapping communication to hardware. Another major advantage is that the set of the transactions can be distributed to processors prior to beginning. This is possible because a processor Pi has to compute Fi,j, j=i,..., n and therefore only the transactions containing the frequent item i are needed.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133616956","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}
Summary form only given. As number and diversity of data, information and resources increase importance of usage of knowledge for their accessing and reuse grows significantly. One of the most useful approaches is based on semantic description of elementary data and pieces of information as well as on relations between them, therefore resulting in ontological approach. Semantics is starting to be used in wider scope than ever, especially for accessing distributed resources. This field is well represented by the Semantic Web, where WWW content can be expressed also in a format that can be read and used by automated tools, thus permitting people and machines to find, share and integrate information more easily. Semantic description of services and user requests can also be used in (semi-)automatic construction of distributed/Grid applications, often in a form of workflows. In order to effectively manage ontologies on the level higher than the local one, e.g., using Virtual Organization approach some sort of knowledge base must be provided.
{"title":"Semantic Approach to Capability and Capacity Computing","authors":"J. Kitowski","doi":"10.1109/ISPDC.2008.67","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.67","url":null,"abstract":"Summary form only given. As number and diversity of data, information and resources increase importance of usage of knowledge for their accessing and reuse grows significantly. One of the most useful approaches is based on semantic description of elementary data and pieces of information as well as on relations between them, therefore resulting in ontological approach. Semantics is starting to be used in wider scope than ever, especially for accessing distributed resources. This field is well represented by the Semantic Web, where WWW content can be expressed also in a format that can be read and used by automated tools, thus permitting people and machines to find, share and integrate information more easily. Semantic description of services and user requests can also be used in (semi-)automatic construction of distributed/Grid applications, often in a form of workflows. In order to effectively manage ontologies on the level higher than the local one, e.g., using Virtual Organization approach some sort of knowledge base must be provided.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128301295","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 proposes a new token-based mutual exclusion algorithm for directed graph topology, with or without cycles. In one of our earlier works, we have introduced a token based algorithm assuming an inverted tree topology. Such a stable, hierarchical topology is quite unrealistic for mobile networks due to frequent link failures and node mobility. The proposed solution overcomes this. Besides maintaining the correctness in terms of liveness and safety, the proposed algorithm ensures fairness in allocating the token amongst the competing processes. The algorithm offers solution paths even in presence of link failures.
{"title":"A Link-failure Resilient Token based Mutual Exclusion Algorithm for Directed Graph Topology","authors":"Sukhendu Kanrar, S. Choudhury, N. Chaki","doi":"10.1109/ISPDC.2008.17","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.17","url":null,"abstract":"This paper proposes a new token-based mutual exclusion algorithm for directed graph topology, with or without cycles. In one of our earlier works, we have introduced a token based algorithm assuming an inverted tree topology. Such a stable, hierarchical topology is quite unrealistic for mobile networks due to frequent link failures and node mobility. The proposed solution overcomes this. Besides maintaining the correctness in terms of liveness and safety, the proposed algorithm ensures fairness in allocating the token amongst the competing processes. The algorithm offers solution paths even in presence of link failures.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132884093","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}
Urban settings are host of diverse networking technologies, from 3G and wired broadband to Wi-Fi, WiMAX, Bluetooth and, more recently, sensor networks. Potentially, these co-existing networks can provide rich inputs to mobile applications that are context aware. To date, the vast majority of context-aware mobile applications are specialized, e.g. healthcare and tourism, and they lack generality and adaptability. It is therefore essential to use a context model that will match the requirements of a large range of future urban networks applications. The question addressed by this paper is to propose an appropriate context model that is general enough to be effective for urban settings, the challenge being to provide the context-aware service to a wide variety of mobile applications across all networks. This paper presents an analysis of existing models, introduces the O+L model (ontologies plus logic) adopted by the authors, the original service architecture that supports the model and its qualitative evaluation.
{"title":"Context Modeling for Urban Mobile Applications","authors":"Romain Williot, D. Grigoras","doi":"10.1109/ISPDC.2008.22","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.22","url":null,"abstract":"Urban settings are host of diverse networking technologies, from 3G and wired broadband to Wi-Fi, WiMAX, Bluetooth and, more recently, sensor networks. Potentially, these co-existing networks can provide rich inputs to mobile applications that are context aware. To date, the vast majority of context-aware mobile applications are specialized, e.g. healthcare and tourism, and they lack generality and adaptability. It is therefore essential to use a context model that will match the requirements of a large range of future urban networks applications. The question addressed by this paper is to propose an appropriate context model that is general enough to be effective for urban settings, the challenge being to provide the context-aware service to a wide variety of mobile applications across all networks. This paper presents an analysis of existing models, introduces the O+L model (ontologies plus logic) adopted by the authors, the original service architecture that supports the model and its qualitative evaluation.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114454503","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}
When running large parallel applications with demands for resources that exceed the capacity the local computing site offers, the deployment in a distributed grid environment may help to satisfy these demands. However, since such an environment is a heterogeneous system by nature, there are some drawbacks that, if not taken into account, are limiting its applicability.First of all, one has to apply a meta-computing or Grid-enabled message-passing library in order to have the ability to route messages to remote sites as well as still being able to exploit fast site-local network facilities.Then, because the inter-site communication usually constitutes the system's bottleneck, appropriate quality of service parameters should be provided and policed for those connections during the application's execution. And finally, the parallel runtime environment of the distributed application should offer service interfaces in order to enable its interaction with Grid middleware. In this paper, we present a new library called ISI whose functionalities meet those requirements in terms of a session layer to be integrated into grid-enabled message-passing implementations.
{"title":"Design and Implementation of a Service-integrated Session Layer for Efficient Message Passing in Grid Computing Environments","authors":"Carsten Clauss, Stefan Lankes, T. Bemmerl","doi":"10.1109/ISPDC.2008.41","DOIUrl":"https://doi.org/10.1109/ISPDC.2008.41","url":null,"abstract":"When running large parallel applications with demands for resources that exceed the capacity the local computing site offers, the deployment in a distributed grid environment may help to satisfy these demands. However, since such an environment is a heterogeneous system by nature, there are some drawbacks that, if not taken into account, are limiting its applicability.First of all, one has to apply a meta-computing or Grid-enabled message-passing library in order to have the ability to route messages to remote sites as well as still being able to exploit fast site-local network facilities.Then, because the inter-site communication usually constitutes the system's bottleneck, appropriate quality of service parameters should be provided and policed for those connections during the application's execution. And finally, the parallel runtime environment of the distributed application should offer service interfaces in order to enable its interaction with Grid middleware. In this paper, we present a new library called ISI whose functionalities meet those requirements in terms of a session layer to be integrated into grid-enabled message-passing implementations.","PeriodicalId":125975,"journal":{"name":"2008 International Symposium on Parallel and Distributed Computing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2008-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122680528","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：