| 1. |
|
|
| 2. |
|
|
| 3. |
- de Blanche, Andreas, 1975, et al.
(författare)
-
Dual Core Efficiency for Engineering Simulation Applications
- 2008
-
Ingår i: International Conference on Parallel and Distributed Processing Techniques and Applications, Las Vegas, NV, USA, 14-17 July 2008. ; , s. 888-894
-
Konferensbidrag (refereegranskat)abstract
- With the event of multi-core processors the parallel execution of simulation applications has resulted in new problems and possibilities in resource usage in high performance computing (HPC). In this paper we have investigated the impact of execution of engineering applications utilizing one and two cores in an Intel Core 2 Duo based Linux cluster. In engineering industry the number of licenses puts practical and economical constraints on the maximum number of processes. Consequently the issue of how to distribute a given number of processes over the compute nodes in a HPC resource becomes very important. When distributing the application over multiple nodes we found that having N processes on N computer nodes, only using one core on each node, is significantly faster than running N processes on N cores in N/2 computer nodes. Only in one case out of 32 it was beneficial to use both cores. The “one compute node – one simulation process” approach gave an average cost efficiency increase of 16.5%, and for several sub-cases it is actually costbeneficial to run on more nodes than fewer, which decreases the overall run time.
|
|
| 4. |
- de Blanche, Andreas, 1975-, et al.
(författare)
-
Multicore Clusters for CFD Simulations : Comparative Study of Three CFD-Softwares
- 2012
-
Ingår i: PROCEEDINGS OFTHE 2012 INTERNATIONAL CONFERENCE ONPARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES ANDAPPLICATIONS, PART II. - : CSREA Press. - 1601322275 - 1601322283 ; , s. 855-852
-
Konferensbidrag (refereegranskat)abstract
- Multicore processors have come to stay, fulfill Moore’s law and might very well revolutionize the computer industry. However, we are now in a transitional period before the new programming models, numerical algorithms and general computer architecture have been developed and the software has been rewritten. This paper focuses on the effects multicore based systems have on industrial computational fluid dynamics (CFD) simulations. The most significant finding was that five of the models ran faster when only one process was executed on each multicore node instead of two. In these cases the execution time was increased by between 6.5% and 64% with a median increase of 10% when utilizing both cores.
|
|
| 5. |
- Namaki, Nima, 1975-, et al.
(författare)
-
A Tool for Processor Dependency Characterization of HPC Applications
- 2009
-
Ingår i: Proceedings for the HPC Asia & APAN 2009. - Hsinchu, Taiwan : National Center for High-Performance Computing. - 9789868522800 ; , s. 70-76
-
Konferensbidrag (refereegranskat)abstract
- In this paper we have implemented and verified Cpugen, a tool for characterization of processor resource utilization of HPC applications .Toward this end we implemented Cpugen, an application with good accuracy for processor load generation. Cpugen was verified through three different phases of passive, active and real world application measurements. The measurement results show that our implemented method is a viable option for non-intrusive, stable and robust load generation. The error range for all generated target loads are between 0.00% minimum and 1.04% maximum, with a median deviation of 0.11%. We can conclude that the method utilized in this investigation provides the ability to generate stable and robust processor load.
|
|
| 6. |
- Namaki, Nima, 1975-, et al.
(författare)
-
Exhausted Dominated Performance : Basic Proof of Concept
- 2010
-
Ingår i: International conference on Parallel and Distributed Processing Techniques and Applications. - Las Vegas : CSREA. - 1601321562 - 1601321570 ; , s. 63-67
-
Konferensbidrag (refereegranskat)
|
|
| 7. |
- Namaki, Nima, 1975-, et al.
(författare)
-
Exhaustion dominated performance : a first attempt
- 2009
-
Ingår i: SAC '09. - New York, NY, USA : ACM. - 9781605581668 ; , s. 1011-1012
-
Konferensbidrag (refereegranskat)abstract
- In this paper we present a first attempt to an analytical method to discover and understand how the available resources influence the execution time. Our method is based on a piecewise linear model for dominating execution limitations and black-box observations. We verify this analysis method by a set of real-world experiments. Finally, we conclude that the different effects follow a linear superposition within a certain range.
|
|
| 8. |
|
|
| 9. |
|
|
| 10. |
|
|
| 11. |
- Namaki, Nima, 1975
(författare)
-
Exhaustion Dominated Performance: Methodology, Tools and Empirical Experiments
- 2008
-
Licentiatavhandling (övrigt vetenskapligt/konstnärligt)abstract
- Problem with application sensitivity to insufficient resources in High PerformanceComputing (HPC) clusters has been a longstanding issue of concern for industries. Inthis thesis we propose a method based on a black box approach for characterizationand analysis of engineering simulation applications with respect to available hardwareresources in the situation of resource depletion. The basis of this hypothesis is theexistence of one dominating bottleneck at any given moment during execution time.This method suggests that engineering simulation application’s behavior to exhaustioncan be explained as an approximately linear dependency of execution time for theproblem at hand. Verification of this hypothesis required accurate non-intrusive measurement,precise methods and tools for individual depletion of hardware resources. The resultsof these efforts were methods for depletion of available bandwidth, available RAMmemory, processor capacity of compute nodes and practical guidelines for recognitionof bottlenecks.The method of bandwidth depletion succeeded to decrease the available bandwidth,by generating stable and robust network traffic, with only 0.5% deviation from thedesired target with a maximum variance of approximately 1%.Furthermore, the method for depletion of processor capacity resulted to a successfulto generate artificial CPU load and occupy as well as 90% of the processor capacity.The verification for this tool confirmed the accuracy of 0.11% median deviation. Theerror range for attempts for a target load lay between 0.00% at minimum and 1.04% atmaximum.Our experiments with HPL and Fluent whilst exhausting the RAM memory and theavailable bandwidth of the computer cluster, confirmed that our proposed method foranalysis and recognition of bottlenecks is fully applicable and a viable option forcharacterization in this domain.
|
|
