Inexpensive Green Mini Supercomputer Based on Single Board Computer Cluster

Authors

  • T.K. Priyambodo Department of Computer Science and Electronics, Faculty of Mathematics and Natural Sciences, Universitas Gadjah Mada.
  • A.W. Lisan Department of Computer Science and Electronics, Faculty of Mathematics and Natural Sciences, Universitas Gadjah Mada.
  • M. Riasetiawan Department of Computer Science and Electronics, Faculty of Mathematics and Natural Sciences, Universitas Gadjah Mada.

Keywords:

Green Computing, High Performance Computing, Low Power, Overclocking, Parallel Computer, Raspberry Pi,

Abstract

This work focused on building a cluster computer named Wisanggeni 01 as mini-supercomputer for high performance computing, research, and educational purposes. Wisanggeni 01 was constructed from 33 node Raspberry Pi 2. Wisanggeni 01 runs Rasbian Whezzy OS with MPICH as parallel protocol. The Wisanggeni 01 performance is optimized with overclocking. The Wisanggeni 01 performance was tested with HPL benchmark, temperature test, and power consumption test. The result indicated that the peak performance of Wisanggeni 01 are 6020 MFLOPS with N=59000 at default clock and 9943 MFLOPS with N=55000 when overclocked. Average temperature when idle is 27.1°C – 30.2°C and 31.2°C – 34.6°C when running HPL benchmark. Average temperature at overclocked mode increase 2°C higher. Maximum wattage load of Wisanggeni 01 are 110W at default clock and 125W at overclocked clock. Power consumption used are 56% Raspberry Pi, 31% switch, and 13% cooler – LED.

References

Rajaraman, V., 1999, “Supercomputers”, University Press: Jawaharlal Nehru Centre for Advanced Scientific Research, India.

Raspberry Pi Foundation: Raspberry Pi FAQs. URL http://www.raspberrypi.org/faqs. Accessed Aug 2016

Cox, S. J., Cox, J. T., Boardman, R., Johnston, S., Scott, M., Brien, N., 2013, “Iridis-Pi: a low-cost, compact demonstration cluster”. University of Southampton, Southampton, UK. URL: http://www.southampton.ac.uk/~sjc/raspberrypi/raspberry_pi_iridis_l ego_supercomputer_paper_cox_Jun2013.pdf.

Cloutier, M.F., Paradis, C., Weaver, V.M., 2014, “Design and Analysis of a 32-bit Embedded High-Performance Cluster Optimized for Energy and Performance – Extended Edition”. University of Maine. URL: http://web.eece.maine.edu/~vweaver/projects/picluster/2014_cohpc_cluster_extended.pdf.

Franks, S., Yerby J., 2014, “Creating a Low Cost Supercomputer with Raspberry Pi”. Southern Association for Information Systems Conference, Macon, GA, USA. URL: http://saisconferencemgmt.org/proceedings/2014/FranksYerby.pdf.

Tso, P.T., White, D.R., Jouet. S, Singer, J., Pezaros, D.P., 2013, “The Glasgow Raspberry Pi Cloud: A Scale Model for Cloud Computing Infrastructures”. University of Glasgow, G12 8QQ, UK. URL: http://davidrwhite.co.uk/wpcontent/uploads/2014/09/the_glasgow_raspberry_pi_cloud_a_scale_ model_for_cloud_computing_infrastructures.pdf.

Apache Hadoop. http://hadoop.apache.org/

Kiepert, J., 2013, “RPiCLUSTER: Creating a raspberry pi-based beowulf cluster”. Technical report, Boise State University. URL: http://coen.boisestate.edu/ece/files/2013/05/Creating.a.Raspberry.PiBased.Beowulf.Cluster_v2.pdf

Dongarra, J.J., Luszczek, P., Petitet, A., 2003, “The linpack benchmark :past, present and future”. Concurrency and Computation: Practiceand Experience. URL: http://onlinelibrary.wiley.com/doi/10.1002/cpe.728/epdf

Dongarra, JJ., 2002, “Performance of various computers using standard linear equations software”. Technical Report,University of Tennessee. URL: http://www.netlib.org/benchmark/performance.ps

Amer, A., Balaji, P., Bland, W., Gropp, W., Latham, R., Lu, H., Oden, L., Pena, A., Raffenetti, K., Seo, S., Thakur, R., Zhang, J., 2015, “MPICH Installer’s Guide Version 3.2”, Mathematics and Computer Science Division Argonne National Library. URL: https://www.mpich.org/static/downloads/3.2/mpich-3.2- installguide.pdf

Cox, S., 2013, “Steps to make Raspberry Pi Supercomputer”. University of Southampton

Plaisier, PJ., 2013, “Raspberry Pi Overclocking Guide – Cheap, Fun, and Overclockable (with HWBOT Prime)”. URL: http://hwbot.org/news/9761_raspberry_pi_overclocking_%e2%80%9 3_cheap_fun_and_hwbot_prime_oc_guide, accessed Feb 2016

Hill, M., Marty, M., 2007, “Amdahl’s Law in the Multicore Era”. IBM. URL: http://research.cs.wisc.edu/multifacet/papers/ieeecomputer08 _amdahl_multicore.pdf

Downloads

Published

2018-02-05

How to Cite

Priyambodo, T., Lisan, A., & Riasetiawan, M. (2018). Inexpensive Green Mini Supercomputer Based on Single Board Computer Cluster. Journal of Telecommunication, Electronic and Computer Engineering (JTEC), 10(1-6), 141–145. Retrieved from https://jtec.utem.edu.my/jtec/article/view/3682