Bills

Computational Methods Across the Curriculum. Ivan Babic Introduction to Parallel Programming and Cluster UW PDF

Description
Computational Methods Across the Curriculum Ivan Babic Introduction to Parallel Programming and Cluster UW 2011 Computational Methods Simulations based on models Analysis of large data sets
Categories
Published
of 20
270
Categories
Published
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Transcript
Computational Methods Across the Curriculum Ivan Babic Introduction to Parallel Programming and Cluster UW 2011 Computational Methods Simulations based on models Analysis of large data sets Visualization of large data sets In general techniques that depend on high performance computing gear to be practical Third method of scientific inquiry 20th vs 21st century science; atoms and icebergs Computational Methods Why use simulations? too small (atoms, molecules) too large (galaxies, the universe) too fast (photosynthesis, protein folding) too slow (geological processes, climate change) too complex (blood circulation, weather) too dangerous (toxic materials, nuclear stockpile stability) High Performance Computing Gear Bazaar Cairo BobSCEd Al-salam Year GFLOPS ~3000 Size 32U 16U 8U 13U Moore s law, every 18 months the density of transistors in integrated circuits roughly doubles Flops indicates how many mathematical operations involving decimal fractions the computer can handle in one second. For PCs it is measured in millions of flops (megaflops), For mainframe computers in billions of flops (gigaflops), For super computers in trillions of flops (teraflops) (floating point operations per second (Flops)) ( / (10^6) = 800 mega FLOPS) yotta FLOPS 10^24 zetta FLOPS 10^21 exa FLOPS 10^18 peta FLOPS 10^15 tera FLOPS 10^12 giga FLOPS 10^9 mega FLOPS 10^6 kilo FLOPS 10^3 Moore s Law in Action Bytes 10^0 phone booth KB 10^3 shoe box MB 10^ GB 10^9 pack of gum credit card TB 10^12? Typical amount of RAM in a desktop computer Data Information Knowledge Complete works of Shakespeare - ~ 5 MB Human genome - ~ 1 GB Complete works of Beethoven - ~ 20 GB Medical imaging - ~ 30 GB per scan Library of Congress - ~ 10 TB All US academic libraries - ~ 2 PB Large Hadron Collider - ~1.5 GB/second Increasingly Readings Data... Parallel and Distributed Computing Decomposing large problems into smaller ones, solving the smaller problems, and then reducing those answers to find the answer Domain decomposition Functional decomposition Shared memory systems Message passing systems Natural Sciences Modeling and Simulation Protein folding Earthquakes Phylogenetic reconstruction Genome construction Data Sets Sloan Digital Sky Survey Protein Data Bank Arctic aerial photographs Geographical information systems (GIS) Protein Folding QuickTime ᆰ and a YUV420 codec decompressor are needed to see this picture. Source: Pande Lab, Stanford University Result of an ensemble molecular dynamics simulation (Gromacs) of the villin headpiece Earthquakes QuickTime ᆰ and a Sorenson Video 3 decompressor are needed to see this picture. Source: San Diego Supercomputer Center Climate Change Source: National Center for Atmospheric Research (NCAR) and the University of Colorado's National Snow and Ice Data Center (NSIDC) Humanities Modeling and Simulation Game theory Topic modeling Text analysis Data Sets Library of Congress Project Gutenberg Newspaper morgues Arts Modeling and Simulation Animation and rendering Painting provenance Digital music Data Sets Photograph archives Scanned paintings, sculptures, buildings Digital recordings Is that really a Van Gogh? Source: Christian Science Monitor Vase with 15 Sunflowers Social Sciences Modeling and Simulation Teacher matching Social systems Derivatives analysis Data Sets Census Geographical information systems (GIS) Voting records Transaction records (commercial and civil) What happened to the Anasazi? Source: Jonathan Rauch, The Atlantic Why Do All This? President s Information Technology Advisory Committee:...computational science is one of the most important technical fields of the 21st century... Rising Above The Gathering Storm:...vastly improving K- 12 and undergraduate science and mathematics education... Bio2010:...exposure during the early years of their undergraduate careers will help life science students use current computer methods and learn how to exploit emerging computer technologies as they arise... Future Software Making sense of all that stuff we are collecting [Readings] Data Information Knowledge Visualization, interactive interfaces Grid Science portals, e.g. TeraGrid, Nanohub, Open Science Grid Humanities, Arts, and Social Science portals Hardware Specialized CPUs, e.g. FPGA and graphics chips Cores, cores, and more cores References * This presentation is based on Charlie's presentation Computation across curriculum
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x