Pub Date : 2007-09-17DOI: 10.1109/CLUSTR.2007.4629275
Olivier Beaumont
This is the sixth edition of the International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks (HeteroPar 2007). The response to the call for papers was very good: we received 17 strong submissions from 9 countries including Austria, China, France, Germany, Hungary, Ireland, Japan, Spain and Tunisia. All submitted papers were distributed to four members of the Program Committee for evaluation. The PC members either reviewed the papers themselves, or solicited external reviewers. The reviewing process went quite smoothly, and each paper received at least 3 reviews. The final decisions on acceptance/rejection were based upon the reviews resulting in selection of 9 papers. We congratulate the authors of accepted papers, and we regret that some potentially interesting papers could not be accepted, mainly due to unsatisfactory quality of presentation of the research results. The presentation of the accepted papers is organized in 3 sessions: • Scheduling on Heterogeneous Platforms; • Applications; • New Trends in Heterogeneous Computing. Altogether, the papers cover a broad spectrum of topics presenting new ideas, dedicated algorithms, and tools for efficient use of heterogeneous networks of computers.
{"title":"Message from the HeteroPar 2007 chair","authors":"Olivier Beaumont","doi":"10.1109/CLUSTR.2007.4629275","DOIUrl":"https://doi.org/10.1109/CLUSTR.2007.4629275","url":null,"abstract":"This is the sixth edition of the International Workshop on Algorithms, Models and Tools for Parallel Computing on Heterogeneous Networks (HeteroPar 2007). The response to the call for papers was very good: we received 17 strong submissions from 9 countries including Austria, China, France, Germany, Hungary, Ireland, Japan, Spain and Tunisia. All submitted papers were distributed to four members of the Program Committee for evaluation. The PC members either reviewed the papers themselves, or solicited external reviewers. The reviewing process went quite smoothly, and each paper received at least 3 reviews. The final decisions on acceptance/rejection were based upon the reviews resulting in selection of 9 papers. We congratulate the authors of accepted papers, and we regret that some potentially interesting papers could not be accepted, mainly due to unsatisfactory quality of presentation of the research results. The presentation of the accepted papers is organized in 3 sessions: • Scheduling on Heterogeneous Platforms; • Applications; • New Trends in Heterogeneous Computing. Altogether, the papers cover a broad spectrum of topics presenting new ideas, dedicated algorithms, and tools for efficient use of heterogeneous networks of computers.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2007-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85752342","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 : 2007-09-17DOI: 10.1109/CLUSTR.2007.4629270
T. Mukherjee, G. Varsamopoulos, S. Gupta, S. Rungta
Power-aware and thermal-aware techniques such as power-throttling and workload manipulation have been developed to counter the increasing power density in the current data centers. The basis for any such power-aware and/or thermal-aware technique, however, depends heavily on the equipment's power consumption model assumed. The goal of this paper is to perform power-profiling of different systems-namely, the Dell PowerEdge 1855 and 1955-based on actual power measurements. Gamut (Generic Application eMUlaTor) benchmark, double-precision matrix multiplication, and convolution of two vectors are used for varying the CPU utilization and Disk I/O.
{"title":"Measurement-based power profiling of data center equipment","authors":"T. Mukherjee, G. Varsamopoulos, S. Gupta, S. Rungta","doi":"10.1109/CLUSTR.2007.4629270","DOIUrl":"https://doi.org/10.1109/CLUSTR.2007.4629270","url":null,"abstract":"Power-aware and thermal-aware techniques such as power-throttling and workload manipulation have been developed to counter the increasing power density in the current data centers. The basis for any such power-aware and/or thermal-aware technique, however, depends heavily on the equipment's power consumption model assumed. The goal of this paper is to perform power-profiling of different systems-namely, the Dell PowerEdge 1855 and 1955-based on actual power measurements. Gamut (Generic Application eMUlaTor) benchmark, double-precision matrix multiplication, and convolution of two vectors are used for varying the CPU utilization and Disk I/O.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2007-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87318097","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 : 2007-09-17DOI: 10.1109/CLUSTR.2007.4629272
T. Keller
Summary form only given. Power-related research activities at IBM's Austin Research Lab (ARL) include low power circuitry, power-efficient microprocessor designs, and server systems power measurement and management at many different levels. Researchers are given the opportunity to see their contributions at all levels of product design, since IBM produces its own microprocessors for its System p, i, and z platforms as well as producing System x platforms with other vendors' microprocessors, as well as marketing Tivoli systems management middleware.
只提供摘要形式。在IBM的Austin research Lab (ARL)中,与电源相关的研究活动包括低功耗电路、节能微处理器设计以及许多不同级别的服务器系统电源测量和管理。研究人员有机会在产品设计的各个层面看到他们的贡献,因为IBM为其System p、i和z平台生产自己的微处理器,并使用其他供应商的微处理器生产System x平台,以及销售Tivoli系统管理中间件。
{"title":"Some work in progress at IBM's Austin Research Lab","authors":"T. Keller","doi":"10.1109/CLUSTR.2007.4629272","DOIUrl":"https://doi.org/10.1109/CLUSTR.2007.4629272","url":null,"abstract":"Summary form only given. Power-related research activities at IBM's Austin Research Lab (ARL) include low power circuitry, power-efficient microprocessor designs, and server systems power measurement and management at many different levels. Researchers are given the opportunity to see their contributions at all levels of product design, since IBM produces its own microprocessors for its System p, i, and z platforms as well as producing System x platforms with other vendors' microprocessors, as well as marketing Tivoli systems management middleware.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2007-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79871008","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 : 2007-09-17DOI: 10.1109/CLUSTR.2007.4629269
Michael Jonas, G. Varsamopoulos, S. Gupta
Ongoing research has demonstrated the potential benefits of thermal-aware load placement in data centers to both reduce cooling costs and component failure rates. However, thermal-aware load placement techniques have not been widely deployed in existing data centers. This is mainly because they rely on a thermal map or profile of the data center, the derivation of which is an interruptive process to the data center operation. We propose a noninvasive solution of producing a thermal map; it consists of training a neural network with observed data from actual data center operation. Our results show that gathering the data and selecting a training set is a fast process, while the neural network with no hidden layers achieves the lowest mean squared error.
{"title":"On developing a fast, cost-effective and non-invasive method to derive data center thermal maps","authors":"Michael Jonas, G. Varsamopoulos, S. Gupta","doi":"10.1109/CLUSTR.2007.4629269","DOIUrl":"https://doi.org/10.1109/CLUSTR.2007.4629269","url":null,"abstract":"Ongoing research has demonstrated the potential benefits of thermal-aware load placement in data centers to both reduce cooling costs and component failure rates. However, thermal-aware load placement techniques have not been widely deployed in existing data centers. This is mainly because they rely on a thermal map or profile of the data center, the derivation of which is an interruptive process to the data center operation. We propose a noninvasive solution of producing a thermal map; it consists of training a neural network with observed data from actual data center operation. Our results show that gathering the data and selecting a training set is a fast process, while the neural network with no hidden layers achieves the lowest mean squared error.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2007-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88145819","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 : 2007-09-17DOI: 10.1109/CLUSTR.2007.4629268
R. Raghavendra, Parthasarathy Ranganathan, V. Talwar, Xiaoyun Zhu, Zhikui Wang
Power and cooling are emerging to be key challenges in data center environments. A recent IDC report estimated the worldwide spending on enterprise power and cooling to be more than $30 billion and likely to even surpass spending on new server hardware. Server rated power consumptions have increased by nearly 10X over the past ten years. This has led to increased spending on cooling and power delivery equipment. A 30,000 square feet 10MW data center can need up to five million dollars of cooling infrastructure; similarly, power delivery beyond 60 Amps per rack can pose fundamental issues. The increased power also has implications on electricity costs, with many data centers reporting millions of dollars for annual usage. From an environmental point of view, the Department of Energy's 2007 estimate of 59 billion KWhrs spent in U.S. servers and data centers translates to several million tons of coal consumption and greenhouse gas emission per year. The U.S. Congress recently passed Public Law 109-431, directing the Environmental Protection Agency (EPA) to study enterprise energy use, and several industry consortiums such as the GreenGrid have been formed to address these issues. In addition, power and cooling can also impact compaction and reliability.
{"title":"Motivating co-ordination of power management solutions in data centers","authors":"R. Raghavendra, Parthasarathy Ranganathan, V. Talwar, Xiaoyun Zhu, Zhikui Wang","doi":"10.1109/CLUSTR.2007.4629268","DOIUrl":"https://doi.org/10.1109/CLUSTR.2007.4629268","url":null,"abstract":"Power and cooling are emerging to be key challenges in data center environments. A recent IDC report estimated the worldwide spending on enterprise power and cooling to be more than $30 billion and likely to even surpass spending on new server hardware. Server rated power consumptions have increased by nearly 10X over the past ten years. This has led to increased spending on cooling and power delivery equipment. A 30,000 square feet 10MW data center can need up to five million dollars of cooling infrastructure; similarly, power delivery beyond 60 Amps per rack can pose fundamental issues. The increased power also has implications on electricity costs, with many data centers reporting millions of dollars for annual usage. From an environmental point of view, the Department of Energy's 2007 estimate of 59 billion KWhrs spent in U.S. servers and data centers translates to several million tons of coal consumption and greenhouse gas emission per year. The U.S. Congress recently passed Public Law 109-431, directing the Environmental Protection Agency (EPA) to study enterprise energy use, and several industry consortiums such as the GreenGrid have been formed to address these issues. In addition, power and cooling can also impact compaction and reliability.","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2007-09-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88752892","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 : 2005-01-01DOI: 10.1109/CLUSTR.2005.347095
Ying-chao Zhou, Dan Meng, Xiao-cheng Zhou, Yao Chen
{"title":"Experience Of OS Optimization For Linpack On Dawning4000A","authors":"Ying-chao Zhou, Dan Meng, Xiao-cheng Zhou, Yao Chen","doi":"10.1109/CLUSTR.2005.347095","DOIUrl":"https://doi.org/10.1109/CLUSTR.2005.347095","url":null,"abstract":"","PeriodicalId":92128,"journal":{"name":"Proceedings. IEEE International Conference on Cluster Computing","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2005-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76375405","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137750
Richard S. Wilkins, Xing Du, Robert A. Cochran, M. Popp
Clustering of computer systems to increase application availability has become a common industry practice. While it does increase the availability of applications and their data to users, it does not solve the problem of a disaster (flood, tornado, earthquake, terrorism, civil unrest, etc.) making the entire cluster, and the applications and data it is serving, unavailable. Distance mirroring of an application's data store allows for recovery from disaster but may still result in long periods of unacceptable downtime. This paper describes a method for stretching a standard Wolfpack (Microsoft/sup /spl trade// Cluster Service, MSCS) cluster of Intel architecture servers geographically for disaster tolerance. Server nodes and their storage may be placed at two (or more) distant sites to prevent a single disaster from taking down the entire cluster. Standard cluster semantics and ease of use are maintained using the remote mirroring capabilities of Hewlett-Packard's high-end storage arrays. The design of additional software to control data mirroring behavior when moving or failing-over applications between server nodes is described. Also, software that allows "stretching" the cluster quorum disk between sites in a manner that is transparent to the cluster software and also software for an external arbitrator node that provides rapid recovery from total loss of inter-site communications is described. Flexibility provided by the array's firmware mirroring options (i.e. synchronous or asynchronous I/O mirroring) allows for optimum use of inter-site link bandwidth based on the data safety requirements of individual applications.
{"title":"Disaster tolerant Wolfpack geo-clusters","authors":"Richard S. Wilkins, Xing Du, Robert A. Cochran, M. Popp","doi":"10.1109/CLUSTR.2002.1137750","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137750","url":null,"abstract":"Clustering of computer systems to increase application availability has become a common industry practice. While it does increase the availability of applications and their data to users, it does not solve the problem of a disaster (flood, tornado, earthquake, terrorism, civil unrest, etc.) making the entire cluster, and the applications and data it is serving, unavailable. Distance mirroring of an application's data store allows for recovery from disaster but may still result in long periods of unacceptable downtime. This paper describes a method for stretching a standard Wolfpack (Microsoft/sup /spl trade// Cluster Service, MSCS) cluster of Intel architecture servers geographically for disaster tolerance. Server nodes and their storage may be placed at two (or more) distant sites to prevent a single disaster from taking down the entire cluster. Standard cluster semantics and ease of use are maintained using the remote mirroring capabilities of Hewlett-Packard's high-end storage arrays. The design of additional software to control data mirroring behavior when moving or failing-over applications between server nodes is described. Also, software that allows \"stretching\" the cluster quorum disk between sites in a manner that is transparent to the cluster software and also software for an external arbitrator node that provides rapid recovery from total loss of inter-site communications is described. Flexibility provided by the array's firmware mirroring options (i.e. synchronous or asynchronous I/O mirroring) allows for optimum use of inter-site link bandwidth based on the data safety requirements of individual applications.","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":"75199691","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137769
S. Sistare, Jack A. Test, D. Plauger
We present a new architecture for the integration of distributed resource management systems and parallel run-time environments such as MPI. The architecture solves the long-standing problem of achieving a tight integration between the two in a clean and robust manner that fully enables the functionality of both systems, including resource limit enforcement and accounting. We also present a more uniform command interface to the user, which simplifies the task of running parallel jobs and tools under a resource manager. The architecture is extensible and allows new systems to be incorporated. We describe the properties that a resource management system must have to work in this architecture, and find that these are ubiquitous in the resource management world. Using the Sun/spl trade/ Cluster Runtime Environment, we show the generality of the approach by implementing tight integrations with PBS, LSF and Sun Grid Engine software, and we demonstrate the advantages of a tight integration. No modifications or enhancements to these resource management systems were required, which is in marked contrast to ad-hoc approaches which typically require such changes.
{"title":"An architecture for integrated resource management of MPI jobs","authors":"S. Sistare, Jack A. Test, D. Plauger","doi":"10.1109/CLUSTR.2002.1137769","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137769","url":null,"abstract":"We present a new architecture for the integration of distributed resource management systems and parallel run-time environments such as MPI. The architecture solves the long-standing problem of achieving a tight integration between the two in a clean and robust manner that fully enables the functionality of both systems, including resource limit enforcement and accounting. We also present a more uniform command interface to the user, which simplifies the task of running parallel jobs and tools under a resource manager. The architecture is extensible and allows new systems to be incorporated. We describe the properties that a resource management system must have to work in this architecture, and find that these are ubiquitous in the resource management world. Using the Sun/spl trade/ Cluster Runtime Environment, we show the generality of the approach by implementing tight integrations with PBS, LSF and Sun Grid Engine software, and we demonstrate the advantages of a tight integration. No modifications or enhancements to these resource management systems were required, which is in marked contrast to ad-hoc approaches which typically require such changes.","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":"75007259","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137787
Jeff Siegel, P. Lu
A practical problem faced by users of metacomputers and computational grids is: If my computation can move from one system to another, how can I ensure that my data will still be available to my computation? Depending on the level of software, technical, and administrative support available, a data grid or a distributed file system would be reasonable solutions. However, it is not always possible (or practical) to have a diverse group of systems administrators agree to adopt a common infrastructure to support remote data access. Yet, having transparent access to any remote data is an important, practical capability. We have developed the Trellis File System (Trellis FS) to allow programs to access data files on any file system and on any host on a network that can be named by a Secure Copy Locator (SCL) or a Uniform Resource Locator (URL). Without requiring any new protocols or infrastructure, Trellis can be used on practically any POSIX-based system on the Internet. Read access, write access, sparse access, local caching of data, prefetching, and authentication are supported. Trellis is implemented as a user-level C library, which mimics the standard stream I/O functions, and is highly portable. Trellis is not a replacement for traditional file systems or data grids; it provides new capabilities by overlaying on top of other file systems, including grid-based file systems. And, by building upon an already-existing infrastructure (i.e., Secure Shell and Secure Copy), Trellis can be used in situations where a suitable data grid or distributed file system does not yet exist.
{"title":"User-level remote data access in overlay metacomputers","authors":"Jeff Siegel, P. Lu","doi":"10.1109/CLUSTR.2002.1137787","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137787","url":null,"abstract":"A practical problem faced by users of metacomputers and computational grids is: If my computation can move from one system to another, how can I ensure that my data will still be available to my computation? Depending on the level of software, technical, and administrative support available, a data grid or a distributed file system would be reasonable solutions. However, it is not always possible (or practical) to have a diverse group of systems administrators agree to adopt a common infrastructure to support remote data access. Yet, having transparent access to any remote data is an important, practical capability. We have developed the Trellis File System (Trellis FS) to allow programs to access data files on any file system and on any host on a network that can be named by a Secure Copy Locator (SCL) or a Uniform Resource Locator (URL). Without requiring any new protocols or infrastructure, Trellis can be used on practically any POSIX-based system on the Internet. Read access, write access, sparse access, local caching of data, prefetching, and authentication are supported. Trellis is implemented as a user-level C library, which mimics the standard stream I/O functions, and is highly portable. Trellis is not a replacement for traditional file systems or data grids; it provides new capabilities by overlaying on top of other file systems, including grid-based file systems. And, by building upon an already-existing infrastructure (i.e., Secure Shell and Secure Copy), Trellis can be used in situations where a suitable data grid or distributed file system does not yet exist.","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":"81796092","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 : 2002-09-23DOI: 10.1109/CLUSTR.2002.1137786
Zhengting Gan, Y. Alexeev, R. Kendall, M. Gordon
A distributed data parallel full CI program is described The implementation of the FCI algorithm is organized in a combined Cl driven approach With extra computation we were able to avoid redundant communication, and convert the collective communication into more efficient point-to-point communication. The network performance is further optimized by improved DDI library. Examples show very good speedup performance on 16 node PC clusters. The application of the code is also demonstrated.
{"title":"A distributed data implementation of parallel full CI program","authors":"Zhengting Gan, Y. Alexeev, R. Kendall, M. Gordon","doi":"10.1109/CLUSTR.2002.1137786","DOIUrl":"https://doi.org/10.1109/CLUSTR.2002.1137786","url":null,"abstract":"A distributed data parallel full CI program is described The implementation of the FCI algorithm is organized in a combined Cl driven approach With extra computation we were able to avoid redundant communication, and convert the collective communication into more efficient point-to-point communication. The network performance is further optimized by improved DDI library. Examples show very good speedup performance on 16 node PC clusters. The application of the code is also demonstrated.","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":"78934633","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}