Skyline queries have received considerable attention in the database community. The goal is to retrieve all records in a database that have the property that no other record is better according to all of a given set of criteria. While this problem has been well studied in the computational geometry literature, the solution of this problem in the database context requires techniques designed particularly to handle large amounts of data. In this paper, we show that parallel computing is an effective method to speed up the answering of skyline queries on large data sets. We also propose to preprocess the set of data points to quickly answer subsequent skyline queries on any subset of the dimensions.
{"title":"Parallel Computation of Skyline Queries","authors":"Adan Cosgaya-Lozano, A. Rau-Chaplin, N. Zeh","doi":"10.1109/HPCS.2007.25","DOIUrl":"https://doi.org/10.1109/HPCS.2007.25","url":null,"abstract":"Skyline queries have received considerable attention in the database community. The goal is to retrieve all records in a database that have the property that no other record is better according to all of a given set of criteria. While this problem has been well studied in the computational geometry literature, the solution of this problem in the database context requires techniques designed particularly to handle large amounts of data. In this paper, we show that parallel computing is an effective method to speed up the answering of skyline queries on large data sets. We also propose to preprocess the set of data points to quickly answer subsequent skyline queries on any subset of the dimensions.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125531449","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}
B. Maniymaran, Muthucumaru Maheswaran, Yuanyuan Gao
In this paper, we propose simple and practical strategies to improve the trustworthiness of network positioning schemes. In particular, our strategies make network positioning immune to non-random perturbations such as denial-of-service attacks and localized network congestion. Additionally, we studied the overhead generated by existing network positioning algorithms and propose an algorithm that results in low overhead while retaining very high accuracies. We performed extensive simulations and implementations on PlanetLab to examine the performance trade-offs.
{"title":"Benefits of Clustering in Landmark-Aided Positioning Algorithms","authors":"B. Maniymaran, Muthucumaru Maheswaran, Yuanyuan Gao","doi":"10.1109/HPCS.2007.10","DOIUrl":"https://doi.org/10.1109/HPCS.2007.10","url":null,"abstract":"In this paper, we propose simple and practical strategies to improve the trustworthiness of network positioning schemes. In particular, our strategies make network positioning immune to non-random perturbations such as denial-of-service attacks and localized network congestion. Additionally, we studied the overhead generated by existing network positioning algorithms and propose an algorithm that results in low overhead while retaining very high accuracies. We performed extensive simulations and implementations on PlanetLab to examine the performance trade-offs.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"135 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122490909","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}
Discusses a method of handling the configuration of kernels and ramdisk images for compute nodes in an OSCAR-integrated way. Providing a kernel and ramdisk which differs from the server allows to handle specific hardware and meet new needs (SSI and virtual clusters).
{"title":"OSCAR KernelPicker: Handling Clients Kernels","authors":"Jean Parpaillon","doi":"10.1109/HPCS.2007.24","DOIUrl":"https://doi.org/10.1109/HPCS.2007.24","url":null,"abstract":"Discusses a method of handling the configuration of kernels and ramdisk images for compute nodes in an OSCAR-integrated way. Providing a kernel and ramdisk which differs from the server allows to handle specific hardware and meet new needs (SSI and virtual clusters).","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"258 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123063258","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}
To more effectively use HPC clusters, co-allocating jobs across multiple clusters becomes an attractive possibility with the primary benefit being reduced turnaround time. This, ultimately, depends on the inter- cluster communication cost. In our previous research, we introduced a co-allocation strategy, MBAS, that made use of two threshold values to control allocation: one for control link saturation and another to control job splitting. In this paper, we examine the performance of MBAS. A simulation study concludes that assigning jobs with different priorities according to their communication patterns, and adjusting the threshold values for link saturation level control and chunk size control in splitting jobs, the MBAS co- allocation strategy can significantly improve both user' satisfaction (in terms of turn around time) and system resource utilization consistently, even for jobs having large communication requirements.
{"title":"An Improved Job Co-Allocation Strategy in Multiple HPC Clusters","authors":"J. Qin, M. Bauer","doi":"10.1109/HPCS.2007.7","DOIUrl":"https://doi.org/10.1109/HPCS.2007.7","url":null,"abstract":"To more effectively use HPC clusters, co-allocating jobs across multiple clusters becomes an attractive possibility with the primary benefit being reduced turnaround time. This, ultimately, depends on the inter- cluster communication cost. In our previous research, we introduced a co-allocation strategy, MBAS, that made use of two threshold values to control allocation: one for control link saturation and another to control job splitting. In this paper, we examine the performance of MBAS. A simulation study concludes that assigning jobs with different priorities according to their communication patterns, and adjusting the threshold values for link saturation level control and chunk size control in splitting jobs, the MBAS co- allocation strategy can significantly improve both user' satisfaction (in terms of turn around time) and system resource utilization consistently, even for jobs having large communication requirements.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130837498","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}
Increasingly, multi-core processors, multi-processor nodes and multi-core, multi-processor nodes are finding their way into computer clusters. Clusters built using such nodes are already quite common and, inevitably, will become more so over time. As with any new technology, however, the potential benefits are seldom as easy to attain as we expect them to be. In this paper, we explore three fundamental issues related to the use of multi-core, multi-processor nodes in compute clusters using MPI: inter-communication (messaging) efficiency, cache effects (in particular processor affinity) and initial process distribution. Based on some initial experiments using a subset of the NAS parallel benchmarks running on a small scale cluster with dual core, dual processor nodes, we report results on the impact of these issues. From these results we attempt to extrapolate some simple, guidelines that are likely to be generally applicable for optimizing MPI code running on clusters with multi- core, multi-processor nodes.
{"title":"On the Programming Impact ofMulti-Core,Multi-Processor Nodes inMPI Clusters","authors":"H. Pourreza, P. Graham","doi":"10.1109/HPCS.2007.23","DOIUrl":"https://doi.org/10.1109/HPCS.2007.23","url":null,"abstract":"Increasingly, multi-core processors, multi-processor nodes and multi-core, multi-processor nodes are finding their way into computer clusters. Clusters built using such nodes are already quite common and, inevitably, will become more so over time. As with any new technology, however, the potential benefits are seldom as easy to attain as we expect them to be. In this paper, we explore three fundamental issues related to the use of multi-core, multi-processor nodes in compute clusters using MPI: inter-communication (messaging) efficiency, cache effects (in particular processor affinity) and initial process distribution. Based on some initial experiments using a subset of the NAS parallel benchmarks running on a small scale cluster with dual core, dual processor nodes, we report results on the impact of these issues. From these results we attempt to extrapolate some simple, guidelines that are likely to be generally applicable for optimizing MPI code running on clusters with multi- core, multi-processor nodes.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"2013 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127411961","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}
One of the challenges in the large-scale simulations required for many molecular systems (such as those of biological interested) is the recording, monitoring and visualization of configurational information from molecular dynamics trajectories spanning millions, and sometimes billions, of timesteps. A detailed record of instantaneous configurations along the full trajectory can quickly become unmanageable. In this paper we will describe an alternative approach that utilizes time coarse-graining, where configurations averaged over trajectory segments are used to follow the detailed molecular behaviour of a system over multiple-nanosecond simulations. We will demonstrate that the sensitivity of structural measures (order parameters) can be dramatically enhanced when applied to averaged configurations. We then specifically discuss the successful application of this approach to molecular dynamics simulations of crystal growth.
{"title":"Utilizing Averaged Configuations from Molecular Dynamics Simulation Trajectories","authors":"P. Kusalik, K. Gillis, J. Vatamanu","doi":"10.1109/HPCS.2007.34","DOIUrl":"https://doi.org/10.1109/HPCS.2007.34","url":null,"abstract":"One of the challenges in the large-scale simulations required for many molecular systems (such as those of biological interested) is the recording, monitoring and visualization of configurational information from molecular dynamics trajectories spanning millions, and sometimes billions, of timesteps. A detailed record of instantaneous configurations along the full trajectory can quickly become unmanageable. In this paper we will describe an alternative approach that utilizes time coarse-graining, where configurations averaged over trajectory segments are used to follow the detailed molecular behaviour of a system over multiple-nanosecond simulations. We will demonstrate that the sensitivity of structural measures (order parameters) can be dramatically enhanced when applied to averaged configurations. We then specifically discuss the successful application of this approach to molecular dynamics simulations of crystal growth.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133863010","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. Mateescu, W. Podaima, A. Charbonneau, R. Impey, M. Viswanathan, A. Agarwal, P. Armstrong, R. Desmarais, I. Gable, S. Popov, Simon Ramage, R. Sobie, D. Vanderster, D. Quesnel
GridXl is a computational grid designed and built to link resources at a number of research institutions across Canada. Building upon the experience of designing, deploying and operating the first generation of GridXl, we have designed a second-generation, Web-services-based, computational grid. The second generation of GridXl leverages the Web services resource framework, implemented by the Globus Toolkit version 4. The value added by GridXl includes metascheduling, file staging, resource registry and resource monitoring.
{"title":"The GridX1 computational Grid: from a set of service-specific protocols to a service-oriented approach","authors":"G. Mateescu, W. Podaima, A. Charbonneau, R. Impey, M. Viswanathan, A. Agarwal, P. Armstrong, R. Desmarais, I. Gable, S. Popov, Simon Ramage, R. Sobie, D. Vanderster, D. Quesnel","doi":"10.1109/HPCS.2007.31","DOIUrl":"https://doi.org/10.1109/HPCS.2007.31","url":null,"abstract":"GridXl is a computational grid designed and built to link resources at a number of research institutions across Canada. Building upon the experience of designing, deploying and operating the first generation of GridXl, we have designed a second-generation, Web-services-based, computational grid. The second generation of GridXl leverages the Web services resource framework, implemented by the Globus Toolkit version 4. The value added by GridXl includes metascheduling, file staging, resource registry and resource monitoring.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"142 4","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114030757","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 presents an agent based system to increase the life time of node in Wireless Sensor Networks. In wireless sensor network nodes deployment of nodes is random and on large scale. This kind of deployment gives birth to massive sensory data which is redundant in nature. Routing of such kind of unnecessary data not only saturates network resources, but also consumes immense nodes energy. We unadulterated our efforts to enhance the node life time in sensor network by introducing mobile agents. Mobile agents are used to reduce the communication cost, especially over low bandwidth links, by moving the processing function to the data rather than bringing the data to a central processor (sink). Toward this end, we propose our agent based directed diffusion approach. Furthermore, to have better understanding in evaluating the performance of both approaches, we present detailed analytical model of data dissemination for both. The results of our simulation show that agent based directed diffusion provides better performance than directed diffusion in terms of energy consumption and bandwidth saturation.
{"title":"Data Dissemination in Wireless Sensor Networks Using Software Agents","authors":"H. Malik, E. Shakshuki","doi":"10.1109/HPCS.2007.13","DOIUrl":"https://doi.org/10.1109/HPCS.2007.13","url":null,"abstract":"This paper presents an agent based system to increase the life time of node in Wireless Sensor Networks. In wireless sensor network nodes deployment of nodes is random and on large scale. This kind of deployment gives birth to massive sensory data which is redundant in nature. Routing of such kind of unnecessary data not only saturates network resources, but also consumes immense nodes energy. We unadulterated our efforts to enhance the node life time in sensor network by introducing mobile agents. Mobile agents are used to reduce the communication cost, especially over low bandwidth links, by moving the processing function to the data rather than bringing the data to a central processor (sink). Toward this end, we propose our agent based directed diffusion approach. Furthermore, to have better understanding in evaluating the performance of both approaches, we present detailed analytical model of data dissemination for both. The results of our simulation show that agent based directed diffusion provides better performance than directed diffusion in terms of energy consumption and bandwidth saturation.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122851897","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}
Lichun Zhu, R. Kent, A. Aggarwal, Peiris Viranthi, Quazi Rahman, Tarik Elamsy, Ositadimma Ejelike
User management is a vital problem to be solved when constructing a Webportal for accessing grid resources. Most current approaches treat web and grid users separately. This is inconvenient to the system users. This paper presents a user management framework that combines the two kinds of user management. It also supports Single Sign- On and federated identity service through integration of different technologies such as Shibboleth, MyProxy and PURSe portlets.
{"title":"Construction of a Webportal and User Management Framework for Grid","authors":"Lichun Zhu, R. Kent, A. Aggarwal, Peiris Viranthi, Quazi Rahman, Tarik Elamsy, Ositadimma Ejelike","doi":"10.1109/HPCS.2007.12","DOIUrl":"https://doi.org/10.1109/HPCS.2007.12","url":null,"abstract":"User management is a vital problem to be solved when constructing a Webportal for accessing grid resources. Most current approaches treat web and grid users separately. This is inconvenient to the system users. This paper presents a user management framework that combines the two kinds of user management. It also supports Single Sign- On and federated identity service through integration of different technologies such as Shibboleth, MyProxy and PURSe portlets.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128478215","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}
Many job schedulers for clusters allow system administrators to specify a variety of policy constraints that can be used to implement site objectives. An example of a site objective could be to achieve high system utilization while at the same time giving users fair access to the resources. Policy can include such things as constraints on walltime, number of processors and memory that can be requested, fairshare targets, and privileges to certain users. This paper highlights the importance and uses of job scheduler policy information in job analysis. A suite of job analysis services that make use of job scheduler policy information is described. Using the analysis services, system administrators can determine if the current policy is having the desired effect and adjust policy as appropriate. Users can gain insight into the utilization they are achieving and determine more appropriate job submission strategies that could increase utilization or reduce queuing delays. Examples of the job analysis suite applied to production data from resources part of WestGrid, a high performance computing consortium in Western Canada, are also provided.
{"title":"Policy Based Job Analysis","authors":"Roger Curry, C. Kiddle, R. Simmonds","doi":"10.1109/HPCS.2007.26","DOIUrl":"https://doi.org/10.1109/HPCS.2007.26","url":null,"abstract":"Many job schedulers for clusters allow system administrators to specify a variety of policy constraints that can be used to implement site objectives. An example of a site objective could be to achieve high system utilization while at the same time giving users fair access to the resources. Policy can include such things as constraints on walltime, number of processors and memory that can be requested, fairshare targets, and privileges to certain users. This paper highlights the importance and uses of job scheduler policy information in job analysis. A suite of job analysis services that make use of job scheduler policy information is described. Using the analysis services, system administrators can determine if the current policy is having the desired effect and adjust policy as appropriate. Users can gain insight into the utilization they are achieving and determine more appropriate job submission strategies that could increase utilization or reduce queuing delays. Examples of the job analysis suite applied to production data from resources part of WestGrid, a high performance computing consortium in Western Canada, are also provided.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122271294","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：