8,504 Hits in 6.6 sec

Evaluation of Intel Memory Drive Technology Performance for Scientific Applications [article]

Vladimir Mironov, Andrey Kudryavtsev, Yuri Alexeev, Alexander Moskovsky, Igor Kulikov, Igor Chernykh
2018 pre-print
In this paper, we present benchmark data for Intel Memory Drive Technology (IMDT), which is a new generation of Software-defined Memory (SDM) based on Intel ScaleMP collaboration and using 3D XPointTM  ...  We studied IMDT performance for synthetic benchmarks, scientific kernels, and applications.  ...  Fedorov for help with Intel® PARDISO benchmark.  ... 
doi:10.1145/3286475.3286479 arXiv:1811.10220v1 fatcat:murmegx6bndorjgsjsbwwuw2he

Using Intel Optane Devices for In-situ Data Staging in HPC Workflows [article]

Pradeep Subedi, Philip E. Davis, J. J. Villalobos, Ivan Rodero, and Manish Parashar
2018 arXiv   pre-print
We study the performance from two perspectives: 1) Benchmarking of drives using FIO workloads, and 2) Assessing the impact of using Optane over NVMe within the DataSpaces framework for in-memory data staging  ...  Intel recently released the Optane drive, which features 3D XPoint memory technology. This device can be deployed as an SSD or as persistent memory.  ...  Comparing performance of Optane vs NVMe SSD In our evaluation tests for characterizing the read/write throughput of Optane Drives, we used an Intel Xeon node, which has 28 Intel(R) Xeon(R) CPU E5-2660  ... 
arXiv:1807.09651v1 fatcat:w323wni4urdozkbjqk565ijcpq

A performance evaluation of scientific I/O workloads on Flash-based SSDs

Stan Park, Kai Shen
2009 2009 IEEE International Conference on Cluster Computing and Workshops  
This paper provides a trace driven performance evaluation of scientific I/O workloads on SSDs.  ...  We find that SSDs only provide modest performance gains over mechanical hard drives due to the writeintensive nature of many scientific workloads.  ...  Our evaluation finds that SSDs only provide modest performance gains over mechanical hard drives due to the write-intensive nature of many scientific workloads.  ... 
doi:10.1109/clustr.2009.5289148 dblp:conf/cluster/ParkS09 fatcat:nlhoygjwpfeepgrgxemdheiqsa

PACS-CS: a large-scale bandwidth-aware PC cluster for scientific computation

T. Boku, M. Sato, A. Ukawa, D. Takahashi, S. Sumimoto, K. Kumon, T. Moriyama, M. Shimizu
2006 Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06)  
Under the preliminary performance evaluation, we confirmed that the computation part in practical Lattice QCD code will be able to achieve 30% of peak performance, and up to 600 Mbyte/sec of bandwidth  ...  The interconnection network for parallel processing is configured as a multi-dimensional Hyper-Crossbar Network based on trunking of Gigabit Ethernet to support large scale scientific computation with  ...  It shows that the memory bandwidth is one of the critical issues to improve the computation performance for this kind of applications.  ... 
doi:10.1109/ccgrid.2006.78 dblp:conf/ccgrid/BokuSUTSKMS06 fatcat:a3ohs5l4bfegzggupmdoks5bfy

Benchmarking: Performance On Alps And Formosa Clusters

Chih-Wei Hsieh, Chau-Yi Chou, Sheng-HsiuKuo, Tsung-Che Tsai, I-Chen Wu
2012 Zenodo  
This paper presents the benchmarking results and performance evaluation of differentclustersbuilt atthe National Center for High-Performance Computingin Taiwan.  ...  Performance of processor, memory subsystem andinterconnect is a critical factor in the overall performance of high performance computing platforms.  ...  There are also some application programs from different scientific and engineering domains currently running on the clusters for performance evaluation.  ... 
doi:10.5281/zenodo.1056463 fatcat:7g2iq74cfbaltjtirfdhjb56ou

Introduction to Xeon Phi Architecture [chapter]

Rezaur Rahman
2013 Intel® Xeon Phi™ Coprocessor Architecture and Tools  
Technical computing can be defined as the application of mathematical and computational principles to solve engineering and scientific problems.  ...  This chapter summarizes the capabilities and limitations of the Intel Xeon Phi coprocessor, as well as key impact points for software and hardware evaluators who are considering this platform for technical  ...  The motive force driving this evolution has been the quest for the ever higher levels of processor performance necessary for executing the computational tasks that today underlie scientific discoveries  ... 
doi:10.1007/978-1-4302-5927-5_1 fatcat:i6ujf5raq5aypf7eksmvjo6kvq

Coarse Grained Parallelized Scientific Applications on a Cost Efficient Intel Atom Based Cluster

Robin Geyer, Andy Georgi, Wolfgang E. Nagel
2011 Procedia Computer Science  
The results of the measurements are discussed and used to evaluate the effectiveness of the systems in terms of performance, power consumption and costs.  ...  Additionally, synthetic benchmarks are conducted to put the performance of the applications in perspective.  ...  BenchIT is a framework that uses synthetic kernels for evaluating different aspects of system performance.  ... 
doi:10.1016/j.procs.2011.04.216 fatcat:wmrien3r25cqvih2r62ylppcvy

The Performance Comparison of Software as a Service for the First Step Cloud Computing Adoption Processes

Dimas Sasongko, Setiya Nugroho
2019 International Journal of Scientific Research in Computer Science Engineering and Information Technology  
The initial stage of the adoption process for the implementation of cloud computing technology in an institution is evaluating to select the cloud provider.  ...  From the results, we infer that Gmail, Google Drive, and Google has responsive web page performance and can be loaded correctly on devices with limited memory.  ...  The performance evaluation parameter is that the low consumption of CPU utilization is better. 3) Memory Peak Usage (max).  ... 
doi:10.32628/cseit195457 fatcat:lrcoruopnnbg3oaoeco7qfc6vi

Early performance evaluation of AVX for HPC

Pawel Gepner, Victor Gamayunov, David L. Fraser
2011 Procedia Computer Science  
We compare two configurations of Intel CPU based systems (one enabled with AVX and a second without AVX enabled) and present a performance evolution of these two platforms.  ...  implications for HPC.  ...  Acknowledgements We gratefully acknowledge the help and support provided by Jamie Wilcox from Intel EMEA Technical Marketing HPC Lab.  ... 
doi:10.1016/j.procs.2011.04.047 fatcat:2vjnedcdkndexd6w24nrusttka

Performance Evaluation of Massively Parallel Systems Using SPECOMP Suite

Dheya Mustafa
2022 Computers  
Performance analysis plays an essential role in achieving a scalable performance of applications on massively parallel supercomputers equipped with thousands of processors.  ...  We present an extensive evaluation study of the performance peaks and scalability of these two modern architectures using SPEC OMP benchmarks.  ...  Acknowledgments: We would like to thank Argonne National Lab for allowing us to use their machines. Conflicts of Interest: The author declares no conflict of interest.  ... 
doi:10.3390/computers11050075 fatcat:4lcuefuno5fwbdibxht43taw74

CERN openlab: Engaging industry for innovation in the LHC Run 3-4 R&D programme

M Girone, A Purcell, A Di Meglio, F Rademakers, K Gunne, M Pachou, S Pavlou
2017 Journal of Physics, Conference Series  
Two of these updates were especially relevant: (i) the introduction of reserved memory, which enables applications to allocate any amount of memory for RapidIO operations; (ii) the introduction of the  ...  In 2016, the High-Throughput Computing Collaboration (HTCC) continued its studies of three new Intel data-centre technologies: Intel Omni-Path, Intel® Xeon™/FPGA, and Intel® Xeon Phi™.  ...  The main challenges in data analytics for scientific and engineering applications involve technology, integration, and education.  ... 
doi:10.1088/1742-6596/898/7/072049 fatcat:fyacklrz5rcnjihy2hx2pmqt6u

Evaluating High Performance Computing Systems at the Naval Research Laboratory

W. Anderson, J. Osburn, R. Rosenberg, M. Lanzagorta
2007 2007 DoD High Performance Computing Modernization Program Users Group Conference  
shared memory and a 73GB 10K performance computing (HPC) assets.  ...  Each node of the NRL (HPCMP), the Center for Computational Science (CCS) system consists of two Opteron 275 2.2 GHz dual core of the Naval Research Laboratory evaluates new high processors with 8GBs of  ...  Currently SGI has no We wish to thank all of the scientists mentioned announced plans to provide further improvements in the  ... 
doi:10.1109/hpcmp-ugc.2007.31 fatcat:knbuwnzvvzfvlejjib425cqei4

DASH: a Recipe for a Flash-based Data Intensive Supercomputer

Jiahua He, Arun Jagatheesan, Sandeep Gupta, Jeffrey Bennett, Allan Snavely
2010 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis  
DASH uses commodity parts including Intel® X25-E flash drives and distributed shared memory (DSM) software from ScaleMP®.  ...  We present here an overview of the design of DASH, an analysis of its cost efficiency, then a detailed recipe for how we designed and tuned it for high data-performance, lastly show that running data-intensive  ...  We would like to credit our science collaborators Peter Nugent and Janet Jacobsen of the Berkeley Lab for the description about PTF presented in section IVB, and also thank them for sharing with us the  ... 
doi:10.1109/sc.2010.16 dblp:conf/sc/HeJGBS10 fatcat:e4kb3h6utfg2vnf6xyuu3fsf7i

Understanding the Impact of Emerging Non-Volatile Memories on High-Performance, IO-Intensive Computing

Adrian M. Caulfield, Joel Coburn, Todor Mollov, Arup De, Ameen Akel, Jiahua He, Arun Jagatheesan, Rajesh K. Gupta, Allan Snavely, Steven Swanson
2010 2010 ACM/IEEE International Conference for High Performance Computing, Networking, Storage and Analysis  
Finally, paging to fast non-volatile memories is a viable option for some applications, providing an alternative to expensive, powerhungry DRAM for supporting scientific applications with large memory  ...  Emerging storage technologies such as flash memories, phase-change memories, and spin-transfer torque memories are poised to close the enormous performance gap between disk-based storage and main memory  ...  Intel rates the SSDs at 250 MB/s for reads and 170 MB/s for writes giving a theoretical peak bandwidth of 1 GB/s for four drives. The drives have a nominal latency of 75 µs.  ... 
doi:10.1109/sc.2010.56 dblp:conf/sc/CaulfieldCMDAHJGSS10 fatcat:kb66vm67gvatnpxfsff3jirc6i

Hardware Developments I - A Survey Of State-Of-The-Art Hardware And Software

Daniel Borgis, Liang Liang, Leon Petit, Michael Lysaght, Alan O'Cais
2016 Zenodo  
Review of actual hardware and software solutions and recommendations to software vendors  ...  codes to other scientific applications, including those of E-CAM.  ...  on Intel Haswell-EP processor technology and one Fat Node Island based on Intel Westmere-EX processor technology.  ... 
doi:10.5281/zenodo.929532 fatcat:cpuc7mplurcqtkunarlbitvdqu
« Previous Showing results 1 — 15 out of 8,504 results