Pub Date : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137765
T. Sterling
Summary form only given. It is observed that once you've finally made it to the North Pole, no matter what direction your next step takes, you can't go anywhere but South. To some, it appears that Commodity Cluster computing is in somewhat of a similar position. Having firmly established itself as the leading approach to scalable high performance computing, commodity clusters such as Beowulf-class systems have no where to go. Yes, they will continue to track Moore's exponential growth in peak performance and storage capacity, and system software will become incrementally better But in terms of strategic advances, commodity clusters are largely corralled in a paradigm cul-de-sac. In some sense there is a kind of truth in this assessment: the fundamental premise of commodity clustering is invariant; Existing, i.e. predeveloped and standalone, systems (uninformatively referred to as "nodes") are integrated by means of a commercially available interconnect network and supporting software. However the flaw of this defeatist perspective is that the nodes look like more of the same, only faster The future of commodity cluster computing will be defined, not solely by Moore's Law, but by the revolutionary structures embodied by the nodes and the innovative ways in which they will be employed. This plenary presentation will explore the innovations in both hardware and software that are likely to drive the future of commodity cluster computing throughout the rest of this decade and toward the trans-Petaflops performance regime of performance. In so doing, we will examine current projections of device technology to anticipate the performance, capacity, power, size, and cost curves of future commodity clusters. Perhaps of more impact are the changes anticipated in hardware architecture including blade technology, system and SMP on a chip, processor in memory architecture, and anticipated advances in networking including Infiniband and optical switching. As new uses are found for clusters with a rapidly expanding customer base including commercial and business communities, a combination of open source and commercial software tools will be developed for ease of use and high productivity as well as resource management and fault recovery As system scale explodes even for moderate cost systems, the software tools to manage them will take on new responsibilities alleviating much of the burden experienced by today's practitioners. The talk will conclude with a look at some more bizarre possibilities made driven by other market and product trends. The future of clusters can show that even at the North Pole, with the right technology and vision, you can also go straight up.
{"title":"Launching into the future of commodity cluster computing","authors":"T. Sterling","doi":"10.1109/CLUSTR.2002.1137765","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137765","url":null,"abstract":"Summary form only given. It is observed that once you've finally made it to the North Pole, no matter what direction your next step takes, you can't go anywhere but South. To some, it appears that Commodity Cluster computing is in somewhat of a similar position. Having firmly established itself as the leading approach to scalable high performance computing, commodity clusters such as Beowulf-class systems have no where to go. Yes, they will continue to track Moore's exponential growth in peak performance and storage capacity, and system software will become incrementally better But in terms of strategic advances, commodity clusters are largely corralled in a paradigm cul-de-sac. In some sense there is a kind of truth in this assessment: the fundamental premise of commodity clustering is invariant; Existing, i.e. predeveloped and standalone, systems (uninformatively referred to as \"nodes\") are integrated by means of a commercially available interconnect network and supporting software. However the flaw of this defeatist perspective is that the nodes look like more of the same, only faster The future of commodity cluster computing will be defined, not solely by Moore's Law, but by the revolutionary structures embodied by the nodes and the innovative ways in which they will be employed. This plenary presentation will explore the innovations in both hardware and software that are likely to drive the future of commodity cluster computing throughout the rest of this decade and toward the trans-Petaflops performance regime of performance. In so doing, we will examine current projections of device technology to anticipate the performance, capacity, power, size, and cost curves of future commodity clusters. Perhaps of more impact are the changes anticipated in hardware architecture including blade technology, system and SMP on a chip, processor in memory architecture, and anticipated advances in networking including Infiniband and optical switching. As new uses are found for clusters with a rapidly expanding customer base including commercial and business communities, a combination of open source and commercial software tools will be developed for ease of use and high productivity as well as resource management and fault recovery As system scale explodes even for moderate cost systems, the software tools to manage them will take on new responsibilities alleviating much of the burden experienced by today's practitioners. The talk will conclude with a look at some more bizarre possibilities made driven by other market and product trends. The future of clusters can show that even at the North Pole, with the right technology and vision, you can also go straight up.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"90723955","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 : 2002-07-29DOI: 10.1109/CLUSTR.2002.1137773
A. Ching, A. Choudhary, W. Liao, R. Ross, W. Gropp
With the tremendous advances in processor and memory technology, I/O has risen to become the bottleneck in high-performance computing for many applications. The development of parallel file systems has helped to ease the performance gap, but I/O still remains an area needing significant performance improvement. Research has found that noncontiguous I/O access patterns in scientific applications combined with current file system methods, to perform these accesses lead to unacceptable performance for large data sets. To enhance performance of noncontiguous I/O, we have created list I/O, a native version of noncontiguous I/O. We have used the Parallel Virtual File System (PVFS) to implement our ideas. Our research and experimentation shows that list I/O outperforms current noncontiguous I/O access methods in most I/O situations and can substantially enhance the performance of real-world scientific applications.
{"title":"Noncontiguous I/O through PVFS","authors":"A. Ching, A. Choudhary, W. Liao, R. Ross, W. Gropp","doi":"10.1109/CLUSTR.2002.1137773","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137773","url":null,"abstract":"With the tremendous advances in processor and memory technology, I/O has risen to become the bottleneck in high-performance computing for many applications. The development of parallel file systems has helped to ease the performance gap, but I/O still remains an area needing significant performance improvement. Research has found that noncontiguous I/O access patterns in scientific applications combined with current file system methods, to perform these accesses lead to unacceptable performance for large data sets. To enhance performance of noncontiguous I/O, we have created list I/O, a native version of noncontiguous I/O. We have used the Parallel Virtual File System (PVFS) to implement our ideas. Our research and experimentation shows that list I/O outperforms current noncontiguous I/O access methods in most I/O situations and can substantially enhance the performance of real-world scientific applications.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83150850","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 : 2002-01-01DOI: 10.1109/CLUSTR.2002.1137780
Daniel Grosu, Anthony T. Chronopoulos
Computational grids are large scale computing systems composed of geographically distributed resources (computers, storage etc.) owned by self interested agents or organizations. These agents may manipulate the resource allocation algorithm for their own benefit and their selfish behavior may lead to severe performance degradation and poor efficiency. In this paper we investigate the problem of designing protocols for resource allocation involving selfish agents. Solving this kind of problem is the object of mechanism design theory. Using this theory we design a truthful mechanism for solving the static load balancing problem in heterogeneous distributed systems. We prove that by using the optimal allocation algorithm the output function admits a truthful payment scheme satisfying voluntary participation. We derive a protocol that implements our mechanism and present experiments to show its effectiveness.
{"title":"Algorithmic mechanism design for load balancing in distributed systems","authors":"Daniel Grosu, Anthony T. Chronopoulos","doi":"10.1109/CLUSTR.2002.1137780","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137780","url":null,"abstract":"Computational grids are large scale computing systems composed of geographically distributed resources (computers, storage etc.) owned by self interested agents or organizations. These agents may manipulate the resource allocation algorithm for their own benefit and their selfish behavior may lead to severe performance degradation and poor efficiency. In this paper we investigate the problem of designing protocols for resource allocation involving selfish agents. Solving this kind of problem is the object of mechanism design theory. Using this theory we design a truthful mechanism for solving the static load balancing problem in heterogeneous distributed systems. We prove that by using the optimal allocation algorithm the output function admits a truthful payment scheme satisfying voluntary participation. We derive a protocol that implements our mechanism and present experiments to show its effectiveness.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2002-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74122050","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 : 2001-10-08DOI: 10.1109/CLUSTR.2001.959953
D. Gaudet
Summary form only given, as follows. Transmeta's Crusoe x86 processor is capable of running at peak performance with low thermal requirements. This allows for a dense arrangement of CPUs with minimal cooling requirements, while achieving competitive performance per watt. We will discuss various performance results using the Crusoe processor.
{"title":"Dense Computing with Transmeta's Crusoe (invited)","authors":"D. Gaudet","doi":"10.1109/CLUSTR.2001.959953","DOIUrl":"https://doi.org/10.1109/CLUSTR.2001.959953","url":null,"abstract":"Summary form only given, as follows. Transmeta's Crusoe x86 processor is capable of running at peak performance with low thermal requirements. This allows for a dense arrangement of CPUs with minimal cooling requirements, while achieving competitive performance per watt. We will discuss various performance results using the Crusoe processor.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2001-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85421701","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 : 2001-10-08DOI: 10.1109/CLUSTER.2001.10004
D. Becker
Beowulf systems are high performance computers constructed from commodity hardware connected by a private internal network and running an open source operating system infrastructure. This tutorial will cover design, installation, deployment and running Beowulf clusters. It will be focused on the Scyld Beowulf system, and will include specific examples and a complete cluster installation CD-ROM.
{"title":"The Scyld Beowulf System","authors":"D. Becker","doi":"10.1109/CLUSTER.2001.10004","DOIUrl":"https://doi.org/10.1109/CLUSTER.2001.10004","url":null,"abstract":"Beowulf systems are high performance computers constructed from commodity hardware connected by a private internal network and running an open source operating system infrastructure. This tutorial will cover design, installation, deployment and running Beowulf clusters. It will be focused on the Scyld Beowulf system, and will include specific examples and a complete cluster installation CD-ROM.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2001-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72449539","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 : 2001-01-01DOI: 10.1109/CLUSTER.2001.10003
W. Gropp
{"title":"Advanced Cluster Programming with MP","authors":"W. Gropp","doi":"10.1109/CLUSTER.2001.10003","DOIUrl":"https://doi.org/10.1109/CLUSTER.2001.10003","url":null,"abstract":"","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2001-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75712052","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 : 2001-01-01DOI: 10.1109/CLUSTER.2001.10000
S. Oberlin
{"title":"\"Architecture Recapitulates Phylogeny\": How Scalability Requires Specializatio","authors":"S. Oberlin","doi":"10.1109/CLUSTER.2001.10000","DOIUrl":"https://doi.org/10.1109/CLUSTER.2001.10000","url":null,"abstract":"","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2001-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80896311","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}