1,268 Hits in 3.4 sec

COMPaS: a PC-based SMP cluster

M. Sato
1999 IEEE Concurrency  
Our objective is to study the performance characteristics of a PC-based SMP cluster and the programming models and methods for an SMP cluster.  ...  In general, however, an SMP cluster often complicates parallel programming because the programmer must take care of both programming models.  ... 
doi:10.1109/4434.749139 fatcat:jqfubtrkqzahdkwo3hazc4jtkm

Message passing and shared address space parallelism on an SMP cluster

Hongzhang Shan, Jaswinder P. Singh, Leonid Oliker, Rupak Biswas
2003 Parallel Computing  
In this paper, we compare the performance of and the programming effort required for six applications under both programming models on a 32-processor PC-SMP cluster, a platform that is becoming increasingly  ...  A hybrid MPI þ SAS strategy shows only a small performance advantage over pure MPI in some cases. Finally, improved implementations of two MPI collective operations on PC-SMP clusters are presented.  ...  Acknowledgements The work of the first two authors was supported by NSF under grant number ESS-9806751 to Princeton University.  ... 
doi:10.1016/s0167-8191(02)00222-3 fatcat:r3erygznb5fcxbz6ai3phlf3ju

Memory hierarchy considerations for cost-effective cluster computing

Xing Du, Xiaodong Zhang, Zhichun Zhu
2000 IEEE transactions on computers  
The model covers the memory hierarchy of a single SMP, a cluster of workstations/PCs, or a cluster of SMPs by changing various architectural parameters.  ...  We present an analytical model for evaluating the performance impact of memory hierarchies and networks on cluster computing.  ...  The report from Referee B was particularly helpful for us to improve the quality and readability of the paper.  ... 
doi:10.1109/12.869323 fatcat:izc5732aqfg43bgcjxusnyieji

A programming model for GPU-based parallel computing with scalability and abstraction

Balázs Domonkos, Gábor Jakab
2009 Proceedings of the 2009 Spring Conference on Computer Graphics - SCCG '09  
In this paper, we present a multi-level programming model for recent GPU-based high performance computing systems.  ...  mapping with code reusability and virtualization of hardware resources in order to decrease the programming effort.  ...  Acknowledgements This work was supported by the Hungarian National Office for Research and Technology (TECH08A2) and Mediso Medical Imaging Systems.  ... 
doi:10.1145/1980462.1980484 dblp:conf/sccg/DomonkosJ09 fatcat:wkwhshot2ba3tgnddnzgqu2qka

Effective cross-platform, multilevel parallelism via dynamic adaptive execution

W. Ko, M. Yankelevsky, D.S. Nikolopoulos, C.D. Polychronopoulos
2002 Proceedings 16th International Parallel and Distributed Processing Symposium  
Using the NAS parallel benchmarks, we first illustrate the lack of portable performance on stateof-the-art scalable parallel systems despite the use of two portable programming models, MPI and OpenMP.  ...  This algorithm can be used as a rule of thumb for automatic multilevel parallelization. The effectiveness of the approach is demonstrated on the NAS benchmarks running on two architectural platforms.  ...  Table 1 illustrates the results for six NAS benchmarks on two platforms, the Origin2000 and the PC cluster.  ... 
doi:10.1109/ipdps.2002.1016495 dblp:conf/ipps/KoYNP02 fatcat:ge65dyzbvbhzxdt6dhvbagls4a

A nationwide parallel computing environment

Ken Kennedy, Charles F. Bender, John W. D. Connolly, John L. Hennessy, Mary K. Vernon, Larry Smarr
1997 Communications of the ACM  
NCSA's plans focus on experimental computer architectures, high performance user programs, machine independent analyzers for parallel computers and advanced application support that leverage parallel adaptive  ...  Alliance (NCSA) hopes to accelerate the creation of a nationwide parallel computing environment for the US by developing hardware and software that will ease desktop access to the National Technology Grid  ...  Already, the two-or four-processor PC or workstation is no longer a novelty.  ... 
doi:10.1145/265684.265693 fatcat:xtfge7keynabjbyxfmdcbm3kci

Language and Compiler Support for Hybrid-Parallel Programming on SMP Clusters [chapter]

Siegfried Benkner, Viera Sipkova
2002 Lecture Notes in Computer Science  
In this paper we present HPF extensions for clusters of SMPs and their implementation within the VFC compiler.  ...  The main goal of these extensions is to optimize HPF for clusters of SMPs by enhancing the functionality of the mapping mechanisms and by providing the user with high-level means for controlling key aspects  ...  In the following we present an experimental evaluation of the new language extensions and the hybrid parallelization strategy as provided by VFC using two benchmark codes on a Beowulf-type SMP PC cluster  ... 
doi:10.1007/3-540-47847-7_4 fatcat:fmg2qm2xcrdexgmrrxwhitswla

Multicores in Cloud Computing: Research Challenges for Applications

Lizhe Wang, Jie Tao, Gregor Von Laszewski, Holger Marten
2010 Journal of Computers  
A cluster of multicores will be commonly used as an efficient computational platform for high performance computing in the near future.  ...  of multicore for high performance applications in the Cloud environments.  ...  Another approach [16] is to program an SMP cluster is using the OpenMP model with a shared virtual memory for the cluster.  ... 
doi:10.4304/jcp.5.6.958-964 fatcat:w7uw2hf7analhadyn7nqrabvym

Advanced Hybrid MPI/OpenMP Parallelization Paradigms for Nested Loop Algorithms onto Clusters of SMPs [chapter]

Nikolaos Drosinos, Nectarios Koziris
2003 Lecture Notes in Computer Science  
We implement the three variations and perform a number of micro-kernel benchmarks to verify the intuition that the hybrid programming model could potentially exploit the characteristics of an SMP cluster  ...  We conclude that the overall performance for each model is both application and hardware dependent, and propose some directions for the efficiency improvement of the hybrid model.  ...  Ò ½ · ½, ÒÓ ); END FOR Hybrid MPI/OpenMP paradigm The hybrid MPI/OpenMP programming model intuitively matches the characteristics of an SMP cluster, since it allows for a two-level communication pattern  ... 
doi:10.1007/978-3-540-39924-7_30 fatcat:nsensostijb2bixnghpbxpyxau

Accurate Performance Evaluation, Modelling and Prediction of a Message Passing Simulation Code based on Middleware

M. Taufer, T. Stricker
1998 Proceedings of the IEEE/ACM SC98 Conference  
to check the suitability of this application for slow CoPs, SMP CoPs and fast CoPs, three flavors of Clusters of PCs built with off-the-shelf Intel Pentium processors.  ...  The predicted execution times and speedup figures indicate that a well designed cluster of PCs achieves similar if not better performance than the J90 vector processors currently used and that the computational  ...  We thank Carol Beaty of the SGI/CRI and Bruno Löpfe of the ETH  ... 
doi:10.1109/sc.1998.10039 dblp:conf/sc/TauferS98 fatcat:rwvmw7jcifdtxfqv7jj7lhrv7m

The performance analysis of portable parallel programming interface MpC for SDSM and pthread

H. Midorikawa
2005 CCGrid 2005. IEEE International Symposium on Cluster Computing and the Grid, 2005.  
In parallel programming model, MPI seems to have established its position, and OpenMP is extensively investigated as the next standard. However, OpenMP is not so efficient for clusters.  ...  Using OpenMP on clusters causes more performance degradation than using SDSMs directly, because most of the OpenMP implementations for clusters use SDSMs in their under layer.  ...  Nagao for the OpenMP performance measurement.  ... 
doi:10.1109/ccgrid.2005.1558656 dblp:conf/ccgrid/Midorikawa05 fatcat:vkqejlqyhrdw7fo6esaghkhhie

Hierarchical Models and Software Tools for Parallel Programming [chapter]

Massimo Coppola, Martin Schmollinger
2003 Lecture Notes in Computer Science  
A second goal is to investigate if software support has been realized, and what is still missing, in order to exploit the full performance of modern highperformance cluster architectures.  ...  We thus need well-founded models and efficient new tools for hierarchical parallel machines, in order to connect algorithm design and complexity results to high-performance program implementation.  ...  discussion which contributed to improve the quality of the paper.  ... 
doi:10.1007/3-540-36574-5_15 fatcat:mgkm5xi34vanfjx2fkryqfoq4i

Performance of cluster-enabled OpenMP for the SCASH software distributed shared memory system

Y. Ojima, M. Sato, H. Harada, Y. Ishikawa
2003 CCGrid 2003. 3rd IEEE/ACM International Symposium on Cluster Computing and the Grid, 2003. Proceedings.  
• To evaluate performance of Omni/SCASH • To investigate the performance factor depending on the communication performance of networks • To investigate the problem of using a commodity network  ...  operation • Measurement condition -programs are parallelized with OpenMP -1 processor per node is used Evaluation platform • PC cluster "COSMO" version 5.0.1 SCore gcc 2.96(Optimize option  ... 
doi:10.1109/ccgrid.2003.1199400 dblp:conf/ccgrid/OjimaSHI03 fatcat:yoboix3so5ebdadu7j526rsbzm

Cluster Configuration Aided by Simulation [chapter]

Dieter F. Kvasnicka, Helmut Hlavacs, Christoph W. Ueberhuber
2001 Lecture Notes in Computer Science  
The acquisition of PC clusters is often limited by financial restrictions.  ...  In this paper it is demonstrated how to use the newly developed simulation tool Clue to decide which configuration of the components of a cluster yields the best price/performance ratio for a particular  ...  of SMPs, (ii) the development of software for parallel computers which are not yet available, (iii) reproducible performance assessments in environments with constantly changing load characteristics (  ... 
doi:10.1007/3-540-45545-0_33 fatcat:c6ulc6tadjasdkcr53ucijv6ii

Parallel computing to start the millennium

Joel Adams, Chris Nevison, Nan C. Schaller
2000 ACM SIGCSE Bulletin  
We describe the experience of three undergraduate computer science programs offering courses on parallel computing.  ...  Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advant  ...  Future Directions RIT bases much of its parallel course on the use of two SMP machines; Calvin College has started to use multithreading on an SMP as well as MPI on a workstation cluster; and Colgate has  ... 
doi:10.1145/331795.331815 fatcat:5xwuamfuujhwbj7r4wp6m4hbfe
« Previous Showing results 1 — 15 out of 1,268 results