Cluster heterogeneity increases the difficulty of balancing the load across the system nodes. Although the relationship between heterogeneity and load balancing is difficult to describe analytically, in this paper an exhaustive analysis of the effects of this system feature on load balancing algorithms performance is presented. Considering the performed analysis, there are two main challenges that need to be faced when dealing with cluster heterogeneity in load balancing algorithms: one related to the state measurement stage and another to the initiation rule. In this paper techniques to deal with heterogeneity in these two algorithm stages are proposed. Furthermore, suggestions to improve the performance of the rest of algorithm stages in heterogeneous environments are made too
{"title":"Dealing with Heterogeneity in Load Balancing Algorithms","authors":"Marta Beltrán, A. Guzmán, J. L. Bosque","doi":"10.1109/ISPDC.2006.17","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.17","url":null,"abstract":"Cluster heterogeneity increases the difficulty of balancing the load across the system nodes. Although the relationship between heterogeneity and load balancing is difficult to describe analytically, in this paper an exhaustive analysis of the effects of this system feature on load balancing algorithms performance is presented. Considering the performed analysis, there are two main challenges that need to be faced when dealing with cluster heterogeneity in load balancing algorithms: one related to the state measurement stage and another to the initiation rule. In this paper techniques to deal with heterogeneity in these two algorithm stages are proposed. Furthermore, suggestions to improve the performance of the rest of algorithm stages in heterogeneous environments are made too","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115751327","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 a parallel implementation of a Lanczos-based method to solve generalised eigenvalue problems related to the modal computation of arbitrarily shaped waveguides. This efficient implementation is intended for execution in moderate-low cost workstations (2 to 4 processors). The problem under study has several features: the involved matrices are sparse with a certain structure, and all the eigenvalues needed are contained in a given interval. The novel parallel algorithms proposed show excellent speed-up for small number of processors
{"title":"Parallel Implementation in PC Clusters of a Lanczos-based Algorithm for an Electromagnetic Eigenvalue Problem","authors":"M. Bernabeu, M. Taroncher, V. García, A. Vidal","doi":"10.1109/ISPDC.2006.39","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.39","url":null,"abstract":"This paper describes a parallel implementation of a Lanczos-based method to solve generalised eigenvalue problems related to the modal computation of arbitrarily shaped waveguides. This efficient implementation is intended for execution in moderate-low cost workstations (2 to 4 processors). The problem under study has several features: the involved matrices are sparse with a certain structure, and all the eigenvalues needed are contained in a given interval. The novel parallel algorithms proposed show excellent speed-up for small number of processors","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115860672","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}
Recent applications for distributed mobile devices, including multimedia video/audio streaming, typically process streams of incoming data in a regular, predictable way. The behavior of these applications during runtime can be accurately predicted most of the time by analyzing the data to be processed and annotating the stream with the information collected. We introduce an annotation-based approach to power-quality trade-offs and demonstrate its application on CPU frequency scaling during video decoding, for an improved user experience on portable devices. Our experiments show that up to 50% of the power consumed by the CPU during video decoding can be saved with this approach
{"title":"Video Stream Annotations for Energy Trade-offs in Multimedia Applications","authors":"R. Cornea, A. Nicolau, N. Dutt","doi":"10.1109/ISPDC.2006.55","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.55","url":null,"abstract":"Recent applications for distributed mobile devices, including multimedia video/audio streaming, typically process streams of incoming data in a regular, predictable way. The behavior of these applications during runtime can be accurately predicted most of the time by analyzing the data to be processed and annotating the stream with the information collected. We introduce an annotation-based approach to power-quality trade-offs and demonstrate its application on CPU frequency scaling during video decoding, for an improved user experience on portable devices. Our experiments show that up to 50% of the power consumed by the CPU during video decoding can be saved with this approach","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116737743","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}
T. Ellahi, B. Hudzia, L. McDermott, Mohand Tahar Kechadi
Grid and P2P systems enable users to share resources across organization boundaries. A strong security system is an integral component of grid middleware to enable secure resource sharing at a very large scale. P2P grid systems combine approaches from both worlds to share resources at a very large scale. P2P grid systems have a different set of security requirements contrary to the traditional server based centralized grid systems. This paper proposes a decentralized security framework for P2P grid systems. This framework is part of a P2P based data grid middleware DGET (data grid environment and tools)
{"title":"Security Framework for P2P Based Grid Systems","authors":"T. Ellahi, B. Hudzia, L. McDermott, Mohand Tahar Kechadi","doi":"10.1109/ISPDC.2006.43","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.43","url":null,"abstract":"Grid and P2P systems enable users to share resources across organization boundaries. A strong security system is an integral component of grid middleware to enable secure resource sharing at a very large scale. P2P grid systems combine approaches from both worlds to share resources at a very large scale. P2P grid systems have a different set of security requirements contrary to the traditional server based centralized grid systems. This paper proposes a decentralized security framework for P2P grid systems. This framework is part of a P2P based data grid middleware DGET (data grid environment and tools)","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"17 2","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114052265","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}
In this paper a systematic study of bisimilarities on asymmetric chine-processes is carried out. The notion of L-bisimilarities on asymmetric chine-processes is introduced. Twelve distinct L-bisimilarities are derived from all of L-bisimilarities by constructing a bisimulation lattice. For each of these twelve distinct L-bisimilarities, its open version is defined and showed to coincide with it, and then its congruence is presented. Three update laws are proposed and three tau laws are modified. Finally, sound complete equational systems are established for twelve congruences
{"title":"Bisimulation Congruence for Asymmetric chi ^ ne -Calculus","authors":"Farong Zhong, Yuxi Fu, Xiaoju Dong","doi":"10.1109/ISPDC.2006.13","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.13","url":null,"abstract":"In this paper a systematic study of bisimilarities on asymmetric chine-processes is carried out. The notion of L-bisimilarities on asymmetric chine-processes is introduced. Twelve distinct L-bisimilarities are derived from all of L-bisimilarities by constructing a bisimulation lattice. For each of these twelve distinct L-bisimilarities, its open version is defined and showed to coincide with it, and then its congruence is presented. Three update laws are proposed and three tau laws are modified. Finally, sound complete equational systems are established for twelve congruences","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"193 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125233559","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 are characterized by low processing power, limited communication bandwidth, limited storage space and rudimentary input and output systems. Networks of such devices are mainly the medium for data exchange. Dynamic code exchange is a feature that is not supported in this environment even though communication is easily achieved. This paper proposes a flexible class loader for the Java connected limited device configuration (CLDC). Introducing a dynamic remote class loader on the CLDC is necessary in order to efficiently deploy the platform into mobile wireless networks. The proposed solution is compared with other possible solutions and is evaluated in relation to existing systems
{"title":"Dynamic Class Provisioning on Mobile Devices","authors":"L. Petrea, D. Grigoras","doi":"10.1109/ISPDC.2006.22","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.22","url":null,"abstract":"Mobile devices are characterized by low processing power, limited communication bandwidth, limited storage space and rudimentary input and output systems. Networks of such devices are mainly the medium for data exchange. Dynamic code exchange is a feature that is not supported in this environment even though communication is easily achieved. This paper proposes a flexible class loader for the Java connected limited device configuration (CLDC). Introducing a dynamic remote class loader on the CLDC is necessary in order to efficiently deploy the platform into mobile wireless networks. The proposed solution is compared with other possible solutions and is evaluated in relation to existing systems","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133343767","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}
Program execution methods and respective architectural solutions for parallel embedded systems with look-ahead dynamic inter-processor connection setting are discussed in the paper. The look-ahead dynamic connection setting is a novel method, which leads to time transparency of reconfiguration control during program execution. The look-ahead dynamic reconfiguration assumes segmentation of programs into sections, which are executed in parallel with connection setting for subsequent sections. The discussed architectures for the look-ahead dynamic reconfiguration are based on redundant resources such as processor links and link connection switches. The paper compares the efficiency of Strassen matrix multiplication in the systems based on multiple link connection switches, partitioned processor link sets and anticipated link connection setting on request in a single crossbar switch
{"title":"Assessment of Dynamic Look-Ahead Inter-Processor Connection Reconfiguration for Different Control Paradigms","authors":"E. Laskowski, M. Tudruj","doi":"10.1109/ISPDC.2006.10","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.10","url":null,"abstract":"Program execution methods and respective architectural solutions for parallel embedded systems with look-ahead dynamic inter-processor connection setting are discussed in the paper. The look-ahead dynamic connection setting is a novel method, which leads to time transparency of reconfiguration control during program execution. The look-ahead dynamic reconfiguration assumes segmentation of programs into sections, which are executed in parallel with connection setting for subsequent sections. The discussed architectures for the look-ahead dynamic reconfiguration are based on redundant resources such as processor links and link connection switches. The paper compares the efficiency of Strassen matrix multiplication in the systems based on multiple link connection switches, partitioned processor link sets and anticipated link connection setting on request in a single crossbar switch","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128955668","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}
Valérie Fiolet, R. Olejnik, Guillem Lefait, B. Toursel
Although many data mining tasks have been parallelized and can thus be executed on dedicated clusters, few solutions currently exist to solve data mining problems on a grid or a non-specialized network of workstations. The current tendency is to focus on the use of grids and/or desktop grids in order to exploit any available workstations with no considerations of their physical positions. If a grid specific algorithm has some common characteristics with a dedicated-cluster algorithm, many constraints are inherent to the use of the grid. In particular, resource volatility and communications cost reduce the parallelism effectiveness. The DisDaMin project (distributed data mining) revisits the data mining tasks and proposes new exploitable algorithms for grids. The DisDaMin mechanisms first implement a specific fragmentation of the data using clustering methods, and then realize asynchronous collaborative techniques according to the specifics of execution on grids. The use of this fragmentation method makes it possible to carry out optimal local processing on each node, with a minimum of communications. Using this, we introduce the distributed algorithm DICCoop, an adaptation of DIC by Brin et al. (1997). Simulations were performed to prove the efficiency of the proposed mechanisms and are hosted on the French national grid GRID5000 (part of the European CoreGrid). We analyse the impact of the numerous parameters on optimization of parallel efficiency
{"title":"Optimal Grid Exploitation Algorithms for Data Mining","authors":"Valérie Fiolet, R. Olejnik, Guillem Lefait, B. Toursel","doi":"10.1109/ISPDC.2006.36","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.36","url":null,"abstract":"Although many data mining tasks have been parallelized and can thus be executed on dedicated clusters, few solutions currently exist to solve data mining problems on a grid or a non-specialized network of workstations. The current tendency is to focus on the use of grids and/or desktop grids in order to exploit any available workstations with no considerations of their physical positions. If a grid specific algorithm has some common characteristics with a dedicated-cluster algorithm, many constraints are inherent to the use of the grid. In particular, resource volatility and communications cost reduce the parallelism effectiveness. The DisDaMin project (distributed data mining) revisits the data mining tasks and proposes new exploitable algorithms for grids. The DisDaMin mechanisms first implement a specific fragmentation of the data using clustering methods, and then realize asynchronous collaborative techniques according to the specifics of execution on grids. The use of this fragmentation method makes it possible to carry out optimal local processing on each node, with a minimum of communications. Using this, we introduce the distributed algorithm DICCoop, an adaptation of DIC by Brin et al. (1997). Simulations were performed to prove the efficiency of the proposed mechanisms and are hosted on the French national grid GRID5000 (part of the European CoreGrid). We analyse the impact of the numerous parameters on optimization of parallel efficiency","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133477190","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 World Wide Web has become pervasive in many facets of our professional and personal lives during the past 10 years. Today, another Internet revolution is emerging: grid computing. While the Web offers easy access to mostly static information via hypertext, the grid adds another fundamental layer by enabling access to and use of the underlying resources. Based on widely accepted grid and Web services standards, resources including computers, storage, scientific instruments and experiments, applications, data, and middleware services communicate with each other and deliver results back to the user. These resources are part of a single service-oriented architecture, called OGSA, the Open Grid Services Architecture. For the past several years, early adopters in research and industry have been building and operating prototypes of grids for global communities, virtual organizations, and within enterprises
{"title":"D-Grid, an E-Science Framework for German Scientists","authors":"W. Gentzsch","doi":"10.1109/ISPDC.2006.16","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.16","url":null,"abstract":"The World Wide Web has become pervasive in many facets of our professional and personal lives during the past 10 years. Today, another Internet revolution is emerging: grid computing. While the Web offers easy access to mostly static information via hypertext, the grid adds another fundamental layer by enabling access to and use of the underlying resources. Based on widely accepted grid and Web services standards, resources including computers, storage, scientific instruments and experiments, applications, data, and middleware services communicate with each other and deliver results back to the user. These resources are part of a single service-oriented architecture, called OGSA, the Open Grid Services Architecture. For the past several years, early adopters in research and industry have been building and operating prototypes of grids for global communities, virtual organizations, and within enterprises","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134029555","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}
Wireless sensor networks play a key role in monitoring remote or inhospitable physical environments. One of the most important constraints is the energy efficiency problem. Power conservation and power management must be taken into account at all levels of the sensor networks system hierarchy. DPM (dynamic power management) technology has been widely used in sensor networks. In this paper, we propose a new energy-efficient DPM, which is a modified sleep state policy developed by Simunic and Chdrakasan (2001) and combined with optimal geographical density control (OGDC) by Zhang and Hou (2004) to keep a minimal number of sensor nodes in the active mode in wireless sensor networks. Implementing dynamic power management with considering the battery status, probability of event generation and OGDC will reduce the energy consumption and prolong the whole lifetime of the sensor networks
{"title":"An Energy-Efficient Dynamic Power Management in Wireless Sensor Networks","authors":"Chuang Lin, Yanxiang He, N. Xiong","doi":"10.1109/ISPDC.2006.8","DOIUrl":"https://doi.org/10.1109/ISPDC.2006.8","url":null,"abstract":"Wireless sensor networks play a key role in monitoring remote or inhospitable physical environments. One of the most important constraints is the energy efficiency problem. Power conservation and power management must be taken into account at all levels of the sensor networks system hierarchy. DPM (dynamic power management) technology has been widely used in sensor networks. In this paper, we propose a new energy-efficient DPM, which is a modified sleep state policy developed by Simunic and Chdrakasan (2001) and combined with optimal geographical density control (OGDC) by Zhang and Hou (2004) to keep a minimal number of sensor nodes in the active mode in wireless sensor networks. Implementing dynamic power management with considering the battery status, probability of event generation and OGDC will reduce the energy consumption and prolong the whole lifetime of the sensor networks","PeriodicalId":196790,"journal":{"name":"2006 Fifth International Symposium on Parallel and Distributed Computing","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-07-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125030490","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：