Outline. In Situ Data Triage and Visualiza8on

Size: px
Start display at page:

Download "Outline. In Situ Data Triage and Visualiza8on"

Transcription

1 In Situ Data Triage and Visualiza8on Kwan- Liu Ma University of California at Davis Outline In situ data triage and visualiza8on: Issues and strategies Case study: An earthquake simula8on Case study: A turbulent combus8on simula8on Closing remarks and discussion 1

2 In Situ Visualiza8on Reducing/visualizing data in situ as the simula8on is running The most feasible solu8on for extreme- scale data analysis What does it mean to be running in situ? A defini8on for in situ processing: Process the data before it is wriuen to the disk. Two technical direc8ons meet this defini8on: Co- located on a node (techniques that exploit data locality) Concurrent processing (shipping data to dedicated vis/analysis nodes, possibly reducing first) 2

3 Mo8va8on for in situ Monitoring: Want to make sure the calcula8on is running well Data reduc8on: We won t be able to write all this data we must do it in situ And new opportuni8es: we have a chance to do something we haven t before we ll get to see all the data and that is a posi8ve we can capture transient events If we find something interes8ng, we can add extra auen8on to that event steering the analysis Debugging and performance op8miza8on Current state of the art Tightly coupled in situ analysis and visualiza8on that are as scalable as the simula8on Efforts for write once, use many in produc8on se`ng (Paraview and VisIt); problems remain in terms of minimizing memory footprint. Some experience understanding what problems can be solved in situ and what can t. Some efforts in iden8fying interes8ng data in situ. Feature detec8on (e.g. finding a hurricane in a climate code) Embedded as part of I/O pipeline (i.e. advanced concurrent processing) 3

4 Issues With Current Approaches Some analysis requires looking at large 8me windows how will we do this? You may not know what is interes8ng up front. New paradigms for skipping back in 8me. How to obtain interac8ve explora8on in this se`ng? Some vis & analysis rou8nes are fundamentally memory heavy (e.g. derived quan88es) and some are fundamentally compute heavy. How does these fit? Not just solving one problem; we will be doing mul8ple analyses concurrently. Further, aimed at single inves8gator approaches how do we allow large groups of people to analyze the data in situ? (e.g. climate, policy decisions) In Situ Visualization Requirements Integra8on of simula8on and visualiza8on codes Low memory overhead Low computa8onal cost Sharing the domain decomposi8on and data structures Scalable parallel visualiza8on algorithms Memory bandwidth is a key issue in some processing models (e.g. the concurrent model where a dedicated node is processing a large amount of data) Addi8onal requirements for interac8ve monitoring/ steering and different types of visualiza8on 4

5 Run simula8on pipelines end- to- end in parallel Eliminate scalability boulenecks Execute all components on the same processors Enable simula8on- 8me visualiza8on steering Increase sustained flops as the problem size increases Won SC06 HPC Analy8cs Challenge on 1024 processors of Cray XT3 5

6 0.661M 9.92M 31.3M 114M 462M 1.22B Elements 6

7 5/30/11 Case Study II: a Turbulent Combustion Simulation Direct numerical simula8on of turbulent combus8on (up to 2025x1600x400) Visualizing both par8cle data and volume data Using a new highly scalable parallel renderer Visualiza8on takes under 1% of overall 8me Using up to 6,480 processors of the Cray XT5 at NCCS/ORNL The largest, most scalable in situ visualiza8on ever achieved (2009) In Situ Vis is feasible! In Situ Visualiza8on Results 1215x960x240 CH2O/HO2 CH3/OH 7

8 Test Results VisIt Through the libsim library of VisIt, the simula8on can interface with VisIt with minimal modifica8on to the simula8on code. There are two interfaces in libsim: one to drive the VisIt server, and one to hand data to the VisIt server upon request. In the libsim library, a database plug- in uses data access callback func8ons to read data from memory in response to requests form the VisIt server 8

9 ParaView Two ways: Through ParaView s coprocessing library providing a programma8c interface to ParaView Using ADIOS I/O framework avoiding disk I/O That is, by using the ADIOS API for I/O, simula8on data can be wriuen to disk or to a stage area hosted by another parallel applica8on. Summary In situ processing offers ways to more intelligently figure out which data to get off the machine Simula8ons will need to accommodate hard analysis issues. Paradigm shiq to analysis- driven science/ discovery- driven science? There will be a transi8on for what scien8sts can and cannot do in terms of analysis; how we will help/ enable this transi8on? Difficult upcoming conversa8ons with communi8es that are leery of in situ. (e.g. climate) There are many remaining research challenges with in situ processing. 9

10 Final Remarks Visualiza8on is the interface for solving data intensive problems Visualiza8on may be used for valida8on, explora8on, and presenta8on. Visualiza8on should be an integral part of the overall engineering design or scien8fic discovery process. In situ processing is absolutely required as we move towards exascale. Discussion 10

11 US DOE ASCR US DOE SciDAC Program US NSF PetaApps Program US NSF OCI Program US NSF CCF Program US NSF FODAVA Program US NSF HECURA Program US AirForce HP Labs Nokia Research AT&T Research Acknowledgments 11

Large Data Visualization

Large Data Visualization Large Data Visualization Seven Lectures 1. Overview (this one) 2. Scalable parallel rendering algorithms 3. Particle data visualization 4. Vector field visualization 5. Visual analytics techniques for

More information

Oh, Exascale! The effect of emerging architectures on scien1fic discovery. Kenneth Moreland, Sandia Na1onal Laboratories

Oh, Exascale! The effect of emerging architectures on scien1fic discovery. Kenneth Moreland, Sandia Na1onal Laboratories Photos placed in horizontal posi1on with even amount of white space between photos and header Oh, $#*@! Exascale! The effect of emerging architectures on scien1fic discovery Ultrascale Visualiza1on Workshop,

More information

SEDA An architecture for Well Condi6oned, scalable Internet Services

SEDA An architecture for Well Condi6oned, scalable Internet Services SEDA An architecture for Well Condi6oned, scalable Internet Services Ma= Welsh, David Culler, and Eric Brewer University of California, Berkeley Symposium on Operating Systems Principles (SOSP), October

More information

Advanced Concepts for Large Data Visualization. SNU, February 28, 2012

Advanced Concepts for Large Data Visualization. SNU, February 28, 2012 Advanced Concepts for Large Data Visualization SNU, February 28, 2012 Research Interests Scientific Visualization Information Visualization Visual Analytics High Performance Computing User Interface Design

More information

High Performance Data Analytics for Numerical Simulations. Bruno Raffin DataMove

High Performance Data Analytics for Numerical Simulations. Bruno Raffin DataMove High Performance Data Analytics for Numerical Simulations Bruno Raffin DataMove bruno.raffin@inria.fr April 2016 About this Talk HPC for analyzing the results of large scale parallel numerical simulations

More information

Parallel Visualiza,on At TACC

Parallel Visualiza,on At TACC Parallel Visualiza,on At TACC Visualiza,on Problems * With thanks to Sean Ahern for the metaphor Huge problems: Data cannot be moved off system where it is computed Visualiza,on requires equivalent resources

More information

Combinatorial Mathema/cs and Algorithms at Exascale: Challenges and Promising Direc/ons

Combinatorial Mathema/cs and Algorithms at Exascale: Challenges and Promising Direc/ons Combinatorial Mathema/cs and Algorithms at Exascale: Challenges and Promising Direc/ons Assefaw Gebremedhin Purdue University (Star/ng August 2014, Washington State University School of Electrical Engineering

More information

Op#mizing MapReduce for Highly- Distributed Environments

Op#mizing MapReduce for Highly- Distributed Environments Op#mizing MapReduce for Highly- Distributed Environments Abhishek Chandra Associate Professor Department of Computer Science and Engineering University of Minnesota hep://www.cs.umn.edu/~chandra 1 Big

More information

Simulation-time data analysis and I/O acceleration at extreme scale with GLEAN

Simulation-time data analysis and I/O acceleration at extreme scale with GLEAN Simulation-time data analysis and I/O acceleration at extreme scale with GLEAN Venkatram Vishwanath, Mark Hereld and Michael E. Papka Argonne Na

More information

A Classifica*on of Scien*fic Visualiza*on Algorithms for Massive Threading Kenneth Moreland Berk Geveci Kwan- Liu Ma Robert Maynard

A Classifica*on of Scien*fic Visualiza*on Algorithms for Massive Threading Kenneth Moreland Berk Geveci Kwan- Liu Ma Robert Maynard A Classifica*on of Scien*fic Visualiza*on Algorithms for Massive Threading Kenneth Moreland Berk Geveci Kwan- Liu Ma Robert Maynard Sandia Na*onal Laboratories Kitware, Inc. University of California at Davis

More information

7 Ways to Increase Your Produc2vity with Revolu2on R Enterprise 3.0. David Smith, REvolu2on Compu2ng

7 Ways to Increase Your Produc2vity with Revolu2on R Enterprise 3.0. David Smith, REvolu2on Compu2ng 7 Ways to Increase Your Produc2vity with Revolu2on R Enterprise 3.0 David Smith, REvolu2on Compu2ng REvolu2on Compu2ng: The R Company REvolu2on R Free, high- performance binary distribu2on of R REvolu2on

More information

Por$ng Monte Carlo Algorithms to the GPU. Ryan Bergmann UC Berkeley Serpent Users Group Mee$ng 9/20/2012 Madrid, Spain

Por$ng Monte Carlo Algorithms to the GPU. Ryan Bergmann UC Berkeley Serpent Users Group Mee$ng 9/20/2012 Madrid, Spain Por$ng Monte Carlo Algorithms to the GPU Ryan Bergmann UC Berkeley Serpent Users Group Mee$ng 9/20/2012 Madrid, Spain 1 Outline Introduc$on to GPUs Why they are interes$ng How they operate Pros and cons

More information

A Distributed Data- Parallel Execu3on Framework in the Kepler Scien3fic Workflow System

A Distributed Data- Parallel Execu3on Framework in the Kepler Scien3fic Workflow System A Distributed Data- Parallel Execu3on Framework in the Kepler Scien3fic Workflow System Ilkay Al(ntas and Daniel Crawl San Diego Supercomputer Center UC San Diego Jianwu Wang UMBC WorDS.sdsc.edu Computa3onal

More information

VisIt Libsim. An in-situ visualisation library

VisIt Libsim. An in-situ visualisation library VisIt Libsim. An in-situ visualisation library December 2017 Jean M. Favre, CSCS Outline Motivations In-situ visualization In-situ processing strategies VisIt s libsim library Enable visualization in a

More information

Danesh TaNi & Amit Amritkar

Danesh TaNi & Amit Amritkar GenIDLEST Co- Design Danesh TaNi & Amit Amritkar Collaborators Wu- chun Feng, Paul Sathre, Kaixi Hou, Sriram Chivukula, Hao Wang, Eric de Sturler, Kasia Swirydowicz Virginia Tech AFOSR- BRI Workshop Feb

More information

Today s Class. High Dimensional Data & Dimensionality Reduc8on. Readings for This Week: Today s Class. Scien8fic Data. Misc. Personal Data 2/22/12

Today s Class. High Dimensional Data & Dimensionality Reduc8on. Readings for This Week: Today s Class. Scien8fic Data. Misc. Personal Data 2/22/12 High Dimensional Data & Dimensionality Reduc8on Readings for This Week: Graphical Histories for Visualiza8on: Suppor8ng Analysis, Communica8on, and Evalua8on, Jeffrey Heer, Jock D. Mackinlay, Chris Stolte,

More information

Hobbes: Composi,on and Virtualiza,on as the Founda,ons of an Extreme- Scale OS/R

Hobbes: Composi,on and Virtualiza,on as the Founda,ons of an Extreme- Scale OS/R Hobbes: Composi,on and Virtualiza,on as the Founda,ons of an Extreme- Scale OS/R Ron Brightwell, Ron Oldfield Sandia Na,onal Laboratories Arthur B. Maccabe, David E. Bernholdt Oak Ridge Na,onal Laboratory

More information

Sec$on 4: Parallel Algorithms. Michelle Ku8el

Sec$on 4: Parallel Algorithms. Michelle Ku8el Sec$on 4: Parallel Algorithms Michelle Ku8el mku8el@cs.uct.ac.za The DAG, or cost graph A program execu$on using fork and join can be seen as a DAG (directed acyclic graph) Nodes: Pieces of work Edges:

More information

Metadata Zoo Dataset Metadata Rebecca Koskela Execu4ve Director, DataONE

Metadata Zoo Dataset Metadata Rebecca Koskela Execu4ve Director, DataONE Metadata Zoo Dataset Metadata Rebecca Koskela Execu4ve Director, DataONE eurocris September 9, 2013 Outline Data Challenges Metadata Solu=on DataONE addressing the Data Challenge Enabling Scien=fic Discovery

More information

Tangible Visualiza.on. Andy Wu Synaesthe.c Media Lab GVU Center Georgia Ins.tute of Technology

Tangible Visualiza.on. Andy Wu Synaesthe.c Media Lab GVU Center Georgia Ins.tute of Technology Tangible Visualiza.on Andy Wu Synaesthe.c Media Lab GVU Center Georgia Ins.tute of Technology Introduc.on Informa.on Visualiza.on (Infovis) is the study of the visual representa.on of complex informa.on,

More information

Collaborative data-driven science. Collaborative data-driven science

Collaborative data-driven science. Collaborative data-driven science Alex Szalay ! Started with the SDSS SkyServer! Built very quickly in 2001! Goal: instant access to rich content! Idea: bring the analysis to the data! Interac

More information

ECP Alpine: Algorithms and Infrastructure for In Situ Visualization and Analysis

ECP Alpine: Algorithms and Infrastructure for In Situ Visualization and Analysis ECP Alpine: Algorithms and Infrastructure for In Situ Visualization and Analysis Presented By: Matt Larsen LLNL-PRES-731545 This work was performed under the auspices of the U.S. Department of Energy by

More information

COL 380: Introduc1on to Parallel & Distributed Programming. Lecture 1 Course Overview + Introduc1on to Concurrency. Subodh Sharma

COL 380: Introduc1on to Parallel & Distributed Programming. Lecture 1 Course Overview + Introduc1on to Concurrency. Subodh Sharma COL 380: Introduc1on to Parallel & Distributed Programming Lecture 1 Course Overview + Introduc1on to Concurrency Subodh Sharma Indian Ins1tute of Technology Delhi Credits Material derived from Peter Pacheco:

More information

DataONE Cyberinfrastructure. Ma# Jones Dave Vieglais Bruce Wilson

DataONE Cyberinfrastructure. Ma# Jones Dave Vieglais Bruce Wilson DataONE Cyberinfrastructure Ma# Jones Dave Vieglais Bruce Wilson Foremost a Federa9on Member Nodes (MNs) Heart of the federa9on Harness the power of local cura9on Coordina9ng Nodes (CNs) Services to link

More information

Introduc)on to Informa)on Visualiza)on

Introduc)on to Informa)on Visualiza)on Introduc)on to Informa)on Visualiza)on Seeing the Science with Visualiza)on Raw Data 01001101011001 11001010010101 00101010100110 11101101011011 00110010111010 Visualiza(on Applica(on Visualiza)on on

More information

Tightly Integrated: Mike Cormier Bill Thackrey. Achieving Fast Time to Value with Splunk. Managing Directors Splunk Architects Concanon LLC

Tightly Integrated: Mike Cormier Bill Thackrey. Achieving Fast Time to Value with Splunk. Managing Directors Splunk Architects Concanon LLC Copyright 2014 Splunk Inc. Tightly Integrated: Achieving Fast Time to Value with Splunk Mike Cormier Bill Thackrey Managing Directors Splunk Cer@fied Architects Concanon LLC Disclaimer During the course

More information

Enabling Scalable Data Analysis for Large Computa9onal Structural Biology Datasets on Distributed Memory Systems

Enabling Scalable Data Analysis for Large Computa9onal Structural Biology Datasets on Distributed Memory Systems Enabling Scalable Data Analysis for Large Computa9onal Structural Biology Datasets on Distributed Memory Systems Michela Taufer Global Compu9ng Laboratory Computer and Informa9on Sciences University of

More information

Monitoring & Analy.cs Working Group Ini.a.ve PoC Setup & Guidelines

Monitoring & Analy.cs Working Group Ini.a.ve PoC Setup & Guidelines Monitoring & Analy.cs Working Group Ini.a.ve PoC Setup & Guidelines Copyright 2017 Open Networking User Group. All Rights Reserved Confiden@al Not For Distribu@on Outline ONUG PoC Right Stuff Innova@on

More information

ECSE 425 Lecture 1: Course Introduc5on Bre9 H. Meyer

ECSE 425 Lecture 1: Course Introduc5on Bre9 H. Meyer ECSE 425 Lecture 1: Course Introduc5on 2011 Bre9 H. Meyer Staff Instructor: Bre9 H. Meyer, Professor of ECE Email: bre9 dot meyer at mcgill.ca Phone: 514-398- 4210 Office: McConnell 525 OHs: M 14h00-15h00;

More information

TerraSwarm. A Machine Learning and Op0miza0on Toolkit for the Swarm. Ilge Akkaya, Shuhei Emoto, Edward A. Lee. University of California, Berkeley

TerraSwarm. A Machine Learning and Op0miza0on Toolkit for the Swarm. Ilge Akkaya, Shuhei Emoto, Edward A. Lee. University of California, Berkeley TerraSwarm A Machine Learning and Op0miza0on Toolkit for the Swarm Ilge Akkaya, Shuhei Emoto, Edward A. Lee University of California, Berkeley TerraSwarm Tools Telecon 17 November 2014 Sponsored by the

More information

Energy- Aware Time Change Detec4on Using Synthe4c Aperture Radar On High- Performance Heterogeneous Architectures: A DDDAS Approach

Energy- Aware Time Change Detec4on Using Synthe4c Aperture Radar On High- Performance Heterogeneous Architectures: A DDDAS Approach Energy- Aware Time Change Detec4on Using Synthe4c Aperture Radar On High- Performance Heterogeneous Architectures: A DDDAS Approach Sanjay Ranka (PI) Sartaj Sahni (Co- PI) Mark Schmalz (Co- PI) University

More information

Big Data, Big Compute, Big Interac3on Machines for Future Biology. Rick Stevens. Argonne Na3onal Laboratory The University of Chicago

Big Data, Big Compute, Big Interac3on Machines for Future Biology. Rick Stevens. Argonne Na3onal Laboratory The University of Chicago Assembly Annota3on Modeling Design Big Data, Big Compute, Big Interac3on Machines for Future Biology Rick Stevens stevens@anl.gov Argonne Na3onal Laboratory The University of Chicago There are no solved

More information

Addressing the System So0ware Challenges for Converged Simula8on and Analysis on Extreme- Scale Systems

Addressing the System So0ware Challenges for Converged Simula8on and Analysis on Extreme- Scale Systems Addressing the System So0ware Challenges for Converged Simula8on and Analysis on Extreme- Scale Systems Ron Brightwell, R&D Manager Scalable System So9ware Department Sandia National Laboratories is a

More information

Web applica*on security for dynamic

Web applica*on security for dynamic Web applica*on security for dynamic languages zane@etsy.com @zanelackey Who am I? Security Engineering Manager @ Etsy Lead AppSec/NetSec/SecEng teams Formerly @ isec Partners Books/presenta*ons primarily

More information

Submitted to: Dr. Sunnie Chung. Presented by: Sonal Deshmukh Jay Upadhyay

Submitted to: Dr. Sunnie Chung. Presented by: Sonal Deshmukh Jay Upadhyay Submitted to: Dr. Sunnie Chung Presented by: Sonal Deshmukh Jay Upadhyay Submitted to: Dr. Sunny Chung Presented by: Sonal Deshmukh Jay Upadhyay What is Apache Survey shows huge popularity spike for Apache

More information

Origin- des*na*on Flow Measurement in High- Speed Networks

Origin- des*na*on Flow Measurement in High- Speed Networks IEEE INFOCOM, 2012 Origin- des*na*on Flow Measurement in High- Speed Networks Tao Li Shigang Chen Yan Qiao Introduc*on (Defini*ons) Origin- des+na+on flow between two routers is the set of packets that

More information

Effec%ve Replica Maintenance for Distributed Storage Systems

Effec%ve Replica Maintenance for Distributed Storage Systems Effec%ve Replica Maintenance for Distributed Storage Systems USENIX NSDI2006 Byung Gon Chun, Frank Dabek, Andreas Haeberlen, Emil Sit, Hakim Weatherspoon, M. Frans Kaashoek, John Kubiatowicz, and Robert

More information

Composing, Reproducing, and Sharing Simula5ons

Composing, Reproducing, and Sharing Simula5ons Composing, Reproducing, and Sharing Simula5ons Daniel Mosse {mosse,childers}@cs.pi

More information

Autonomous Threat Hun?ng With Niddel And Splunk Enterprise Security: Mars Inc. Customer Case Study

Autonomous Threat Hun?ng With Niddel And Splunk Enterprise Security: Mars Inc. Customer Case Study Copyright 2016 Splunk Inc. Autonomous Threat Hun?ng With Niddel And Splunk Enterprise Security: Mars Inc. Customer Case Study Alex Pinto Chief Data Scien?st, Niddel Greg Poniatowski Security Service Area

More information

Automated Verifica/on of I/O Performance. F. Delalondre, M. Baerstchi. EPFL/Blue Brain Project - confiden6al

Automated Verifica/on of I/O Performance. F. Delalondre, M. Baerstchi. EPFL/Blue Brain Project - confiden6al Automated Verifica/on of I/O Performance F. Delalondre, M. Baerstchi Requirements Support Scien6sts Crea6vity Minimize Development 6me Maximize applica6on performance Performance Analysis System Performance

More information

Faster Splunk App Cer=fica=on with Splunk AppInspect

Faster Splunk App Cer=fica=on with Splunk AppInspect Copyright 2016 Splunk Inc. Faster Splunk App Cer=fica=on with Splunk AppInspect Andy Nortrup Product Manager, Splunk Grigori Melnik Director, Product Management, Splunk Disclaimer During the course of this

More information

hashfs Applying Hashing to Op2mize File Systems for Small File Reads

hashfs Applying Hashing to Op2mize File Systems for Small File Reads hashfs Applying Hashing to Op2mize File Systems for Small File Reads Paul Lensing, Dirk Meister, André Brinkmann Paderborn Center for Parallel Compu2ng University of Paderborn Mo2va2on and Problem Design

More information

MPI Performance Analysis Trace Analyzer and Collector

MPI Performance Analysis Trace Analyzer and Collector MPI Performance Analysis Trace Analyzer and Collector Berk ONAT İTÜ Bilişim Enstitüsü 19 Haziran 2012 Outline MPI Performance Analyzing Defini6ons: Profiling Defini6ons: Tracing Intel Trace Analyzer Lab:

More information

Vulnerability Analysis (III): Sta8c Analysis

Vulnerability Analysis (III): Sta8c Analysis Computer Security Course. Vulnerability Analysis (III): Sta8c Analysis Slide credit: Vijay D Silva 1 Efficiency of Symbolic Execu8on 2 A Sta8c Analysis Analogy 3 Syntac8c Analysis 4 Seman8cs- Based Analysis

More information

Tools zur Op+mierung eingebe2eter Mul+core- Systeme. Bernhard Bauer

Tools zur Op+mierung eingebe2eter Mul+core- Systeme. Bernhard Bauer Tools zur Op+mierung eingebe2eter Mul+core- Systeme Bernhard Bauer Agenda Mo+va+on So.ware Engineering & Mul5core Think Parallel Models Added Value Tooling Quo Vadis? The Mul5core Era Moore s Law: The

More information

Embedding System Dynamics in Agent Based Models for Complex Adap;ve Systems

Embedding System Dynamics in Agent Based Models for Complex Adap;ve Systems Embedding System Dynamics in Agent Based Models for Complex Adap;ve Systems Kiyan Ahmadizadeh, Maarika Teose, Carla Gomes, Yrjo Grohn, Steve Ellner, Eoin O Mahony, Becky Smith, Zhao Lu, Becky Mitchell

More information

Welcome and introduction to SU 2

Welcome and introduction to SU 2 Welcome and introduction to SU 2 SU 2 Release Version 2. Workshop Stanford University Tuesday, January 5 th, 23 Dr. F. Palacios and Prof. J. J. Alonso Department of Aeronautics & Astronautics Stanford

More information

Results from the Early Science High Speed Combus:on and Detona:on Project

Results from the Early Science High Speed Combus:on and Detona:on Project Results from the Early Science High Speed Combus:on and Detona:on Project Alexei Khokhlov, University of Chicago Joanna Aus:n, University of Illinois Charles Bacon, Argonne Na:onal Laboratory Andrew Knisely,

More information

A Comparison of GPU Box- Plane Intersec8on Algorithms for Direct Volume Rendering. Chair of Computer Science Prof. Lang University of Cologne, Germany

A Comparison of GPU Box- Plane Intersec8on Algorithms for Direct Volume Rendering. Chair of Computer Science Prof. Lang University of Cologne, Germany A Comparison of GPU Box- Plane Intersec8on Algorithms for Direct Volume Rendering Chair of Computer Science Prof. Lang, Germany Stefan Zellmann (zellmans@uni- koeln.de) Ulrich Lang (lang@uni- koeln.de)

More information

Visualiza(on So-ware and Hardware for In- Silico Brain Research. Stefan Eilemann Visualiza0on Team Lead Blue Brain Project, EPFL

Visualiza(on So-ware and Hardware for In- Silico Brain Research. Stefan Eilemann Visualiza0on Team Lead Blue Brain Project, EPFL Visualiza(on So-ware and Hardware for In- Silico Brain Research Stefan Eilemann Visualiza0on Team Lead Blue Brain Project, EPFL Blue Brain Project / Human Brain Project BBP: Swiss na0onal research project

More information

Scalability in a Real-Time Decision Platform

Scalability in a Real-Time Decision Platform Scalability in a Real-Time Decision Platform Kenny Shi Manager Software Development ebay Inc. A Typical Fraudulent Lis3ng fraud detec3on architecture sync vs. async applica3on publish messaging bus request

More information

China's supercomputer surprises U.S. experts

China's supercomputer surprises U.S. experts China's supercomputer surprises U.S. experts John Markoff Reproduced from THE HINDU, October 31, 2011 Fast forward: A journalist shoots video footage of the data storage system of the Sunway Bluelight

More information

Concurrency-Optimized I/O For Visualizing HPC Simulations: An Approach Using Dedicated I/O Cores

Concurrency-Optimized I/O For Visualizing HPC Simulations: An Approach Using Dedicated I/O Cores Concurrency-Optimized I/O For Visualizing HPC Simulations: An Approach Using Dedicated I/O Cores Ma#hieu Dorier, Franck Cappello, Marc Snir, Bogdan Nicolae, Gabriel Antoniu 4th workshop of the Joint Laboratory

More information

Profiling & Tuning Applica1ons. CUDA Course July István Reguly

Profiling & Tuning Applica1ons. CUDA Course July István Reguly Profiling & Tuning Applica1ons CUDA Course July 21-25 István Reguly Introduc1on Why is my applica1on running slow? Work it out on paper Instrument code Profile it NVIDIA Visual Profiler Works with CUDA,

More information

Asynchronous and Fault-Tolerant Recursive Datalog Evalua9on in Shared-Nothing Engines

Asynchronous and Fault-Tolerant Recursive Datalog Evalua9on in Shared-Nothing Engines Asynchronous and Fault-Tolerant Recursive Datalog Evalua9on in Shared-Nothing Engines Jingjing Wang, Magdalena Balazinska, Daniel Halperin University of Washington Modern Analy>cs Requires Itera>on Graph

More information

CAP5415-Computer Vision Lecture 13-Support Vector Machines for Computer Vision Applica=ons

CAP5415-Computer Vision Lecture 13-Support Vector Machines for Computer Vision Applica=ons CAP5415-Computer Vision Lecture 13-Support Vector Machines for Computer Vision Applica=ons Guest Lecturer: Dr. Boqing Gong Dr. Ulas Bagci bagci@ucf.edu 1 October 14 Reminders Choose your mini-projects

More information

Performance Evaluation of a MongoDB and Hadoop Platform for Scientific Data Analysis

Performance Evaluation of a MongoDB and Hadoop Platform for Scientific Data Analysis Performance Evaluation of a MongoDB and Hadoop Platform for Scientific Data Analysis Elif Dede, Madhusudhan Govindaraju Lavanya Ramakrishnan, Dan Gunter, Shane Canon Department of Computer Science, Binghamton

More information

Bridging the Gap Between High Quality and High Performance for HPC Visualization

Bridging the Gap Between High Quality and High Performance for HPC Visualization Bridging the Gap Between High Quality and High Performance for HPC Visualization Rob Sisneros National Center for Supercomputing Applications University of Illinois at Urbana Champaign Outline Why am I

More information

Integra(ng an open source dynamic river model in hydrology modeling frameworks

Integra(ng an open source dynamic river model in hydrology modeling frameworks Integra(ng an open source dynamic river model in hydrology modeling frameworks Simula(on of Guadalupe and San Antonio River basin during a flood event with 1.3 x 10 5 computa(onal nodes at 100 m resolu(on.

More information

visualisation in-situ

visualisation in-situ Couplage Visualisation/Simulation avec VisIt ou bien visualisation in-situ Jean M. Favre 11-04-2012 Couplage Visualisation/Simulation avec VisIt Résumé ORAP Différentes techniques de couplage calcul/visualisation

More information

Volume Visualiza0on. Today s Class. Grades & Homework feedback on Homework Submission Server

Volume Visualiza0on. Today s Class. Grades & Homework feedback on Homework Submission Server 11/3/14 Volume Visualiza0on h3p://imgur.com/trjonqk h3p://i.imgur.com/zcjc9kp.jpg Today s Class Grades & Homework feedback on Homework Submission Server Everything except HW4 (didn t get to that yet) &

More information

Automated UI tests for Mobile Apps. Sedina Oruc

Automated UI tests for Mobile Apps. Sedina Oruc Automated UI tests for Mobile Apps Sedina Oruc What I ll be covering Ø Basics Ø What are UI tests? Ø The no@on of Emulator and Simulator Ø What are our challenges? Ø PlaForm specific UI tes@ng frameworks

More information

Characterize Energy Impact of Concurrent Network- Intensive Applica=ons on Mobile PlaAorms

Characterize Energy Impact of Concurrent Network- Intensive Applica=ons on Mobile PlaAorms ACM MobiArch 2013 Characterize Energy Impact of Concurrent Network- Intensive Applica=ons on Mobile PlaAorms Zhonghon Ou, Shichao Dong, Jiang Dong, Jukka K. Nurminen, AnH Ylä- Jääski Aalto University,

More information

Op#mizing PGAS overhead in a mul#-locale Chapel implementa#on of CoMD

Op#mizing PGAS overhead in a mul#-locale Chapel implementa#on of CoMD Op#mizing PGAS overhead in a mul#-locale Chapel implementa#on of CoMD Riyaz Haque and David F. Richards This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore

More information

Parallel File Systems. IIT Course: Data- Intensive Compu4ng Guest Lecture Samuel Lang September 20, 2010

Parallel File Systems. IIT Course: Data- Intensive Compu4ng Guest Lecture Samuel Lang September 20, 2010 Parallel File Systems IIT Course: Data- Intensive Compu4ng Guest Lecture Samuel Lang September 20, 2010 What are Parallel File Systems? 2 Parallel File Systems Store applica4on data persistently usually

More information

Crea%ng and U%lizing Linked Open Sta%s%cal Data for the Development of Advanced Analy%cs Services E. Kalampokis, A. Karamanou, A. Nikolov, P.

Crea%ng and U%lizing Linked Open Sta%s%cal Data for the Development of Advanced Analy%cs Services E. Kalampokis, A. Karamanou, A. Nikolov, P. Crea%ng and U%lizing Linked Open Sta%s%cal Data for the Development of Advanced Analy%cs Services E. Kalampokis, A. Karamanou, A. Nikolov, P. Haase, R. Cyganiak, B. Roberts, P. Hermans, E. Tambouris, K.

More information

Op#miza#on of D3Q19 La2ce Boltzmann Kernels for Recent Mul#- and Many-cores Intel Based Systems

Op#miza#on of D3Q19 La2ce Boltzmann Kernels for Recent Mul#- and Many-cores Intel Based Systems Op#miza#on of D3Q19 La2ce Boltzmann Kernels for Recent Mul#- and Many-cores Intel Based Systems Ivan Giro*o 1235, Sebas4ano Fabio Schifano 24 and Federico Toschi 5 1 International Centre of Theoretical

More information

Scaling the Wholesale Interconnect Market. Gastón Cu0gnola Senior Sales Engineer Telco Systems

Scaling the Wholesale Interconnect Market. Gastón Cu0gnola Senior Sales Engineer Telco Systems Host Sponsor Co- Sponsor Scaling the Wholesale Interconnect Market Gastón Cu0gnola Senior Sales Engineer Telco Systems 1 Presenta0on Agenda Status of Wholesale/Interconnect Environments Moving up the curve

More information

HPCSoC Modeling and Simulation Implications

HPCSoC Modeling and Simulation Implications Department Name (View Master > Edit Slide 1) HPCSoC Modeling and Simulation Implications (Sharing three concerns from an academic research user perspective using free, open tools. Solutions left to the

More information

An Introduc+on to HPC Tools Research

An Introduc+on to HPC Tools Research An Introduc+on to HPC Tools Research Karen L. Karavanic Research Scien+st New Mexico Consor+um and Associate Professor of Computer Science Portland State University Karen L. Karavanic An Introduc+on to

More information

CENG505 Advanced Computer Graphics Lecture 1 - Introduction. Instructor: M. Abdullah Bülbül

CENG505 Advanced Computer Graphics Lecture 1 - Introduction. Instructor: M. Abdullah Bülbül CENG505 Advanced Computer Graphics Lecture 1 - Introduction Instructor: M. Abdullah Bülbül 1 What is Computer Graphics? Using computers to generate and display images. 2 Computer Graphics Applica>ons (Where

More information

Hypergraph Sparsifica/on and Its Applica/on to Par//oning

Hypergraph Sparsifica/on and Its Applica/on to Par//oning Hypergraph Sparsifica/on and Its Applica/on to Par//oning Mehmet Deveci 1,3, Kamer Kaya 1, Ümit V. Çatalyürek 1,2 1 Dept. of Biomedical Informa/cs, The Ohio State University 2 Dept. of Electrical & Computer

More information

MPICH: A High-Performance Open-Source MPI Implementation. SC11 Birds of a Feather Session

MPICH: A High-Performance Open-Source MPI Implementation. SC11 Birds of a Feather Session MPICH: A High-Performance Open-Source MPI Implementation SC11 Birds of a Feather Session Schedule MPICH2 status and plans Presenta

More information

Combining Real Time Emula0on of Digital Communica0ons between Distributed Embedded Control Nodes with Real Time Power System Simula0on

Combining Real Time Emula0on of Digital Communica0ons between Distributed Embedded Control Nodes with Real Time Power System Simula0on 1 Combining Real Time Emula0on of Digital Communica0ons between Distributed Embedded Control Nodes with Real Time Power System Simula0on Ziyuan Cai and Ming Yu Electrical and Computer Eng., Florida State

More information

Fusion PIC Code Performance Analysis on the Cori KNL System. T. Koskela*, J. Deslippe*,! K. Raman**, B. Friesen*! *NERSC! ** Intel!

Fusion PIC Code Performance Analysis on the Cori KNL System. T. Koskela*, J. Deslippe*,! K. Raman**, B. Friesen*! *NERSC! ** Intel! Fusion PIC Code Performance Analysis on the Cori KNL System T. Koskela*, J. Deslippe*,! K. Raman**, B. Friesen*! *NERSC! ** Intel! tkoskela@lbl.gov May 18, 2017-1- Outline Introduc3on to magne3c fusion

More information

Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analy;cs 100X Faster and 40X Cheaper

Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analy;cs 100X Faster and 40X Cheaper Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analy;cs 100X Faster and 40X Cheaper November 18, 2015 Super Compu3ng 2015 The Amount of Graph Data is Exploding! Billion+ Edges! 2 Graph Applications

More information

Siemens Sanayi ve Ticaret A.Ş. Şafak Karahan

Siemens Sanayi ve Ticaret A.Ş. Şafak Karahan Interna'onal Brokerage Event Brussels, 26-27/10/2017 Siemens Sanayi ve Ticaret A.Ş. Şafak Karahan Safak.karahan@siemens.com Descrip(on of the Organiza(on Power u'li'es and TSOs Municipali'es and DSOs Renewable

More information

Ar#ficial Intelligence

Ar#ficial Intelligence Ar#ficial Intelligence Advanced Searching Prof Alexiei Dingli Gene#c Algorithms Charles Darwin Genetic Algorithms are good at taking large, potentially huge search spaces and navigating them, looking for

More information

Open-Source Based Solutions for Processing, Preserving, and Presenting Oral Histories

Open-Source Based Solutions for Processing, Preserving, and Presenting Oral Histories Western Washington University From the SelectedWorks of Mark I. Greenberg April 2, 2011 Open-Source Based Solutions for Processing, Preserving, and Presenting Oral Histories Mark I. Greenberg, University

More information

Soft GPGPUs for Embedded FPGAS: An Architectural Evaluation

Soft GPGPUs for Embedded FPGAS: An Architectural Evaluation Soft GPGPUs for Embedded FPGAS: An Architectural Evaluation 2nd International Workshop on Overlay Architectures for FPGAs (OLAF) 2016 Kevin Andryc, Tedy Thomas and Russell Tessier University of Massachusetts

More information

Leveraging Tools and Components from OODT and Apache within Climate Science and the Earth System Grid Federa9on

Leveraging Tools and Components from OODT and Apache within Climate Science and the Earth System Grid Federa9on Leveraging Tools and Components from OODT and Apache within Climate Science and the Earth System Grid Federa9on Luca Cinquini, Dan Crichton, Chris Ma2mann NASA Jet Propulsion Laboratory, California Ins9tute

More information

System Modeling Environment

System Modeling Environment System Modeling Environment Requirements, Architecture and Implementa

More information

FORTISSIMO HPGA HIGH PERFORMANCE GEAR ANALYZER. Ing. Antonio Zippo, PhD PULSAR DYNAMICS - UniMORE

FORTISSIMO HPGA HIGH PERFORMANCE GEAR ANALYZER. Ing. Antonio Zippo, PhD PULSAR DYNAMICS - UniMORE FORTISSIMO HPGA HIGH PERFORMANCE GEAR ANALYZER Ing. Antonio Zippo, PhD PULSAR DYNAMICS - UniMORE PARTNERS Applica?on Expert: Vibra?ons and PowerTrain Lab UNIMORE HPC Expert: CINECA Galileo Supercomputer

More information

OLCF's next- genera0on Spider file system

OLCF's next- genera0on Spider file system OLCF's next- genera0on Spider file system Sarp Oral, PhD File and Storage Systems Team Lead Technology Integra0on Group Oak Ridge Leadership Compu0ng Facility Oak Ridge Na0onal Laboratory April 18, 2013

More information

High-Level Synthesis Creating Custom Circuits from High-Level Code

High-Level Synthesis Creating Custom Circuits from High-Level Code High-Level Synthesis Creating Custom Circuits from High-Level Code Hao Zheng Comp Sci & Eng University of South Florida Exis%ng Design Flow Register-transfer (RT) synthesis - Specify RT structure (muxes,

More information

Lecture 1 Introduc-on

Lecture 1 Introduc-on Lecture 1 Introduc-on What would you get out of this course? Structure of a Compiler Op9miza9on Example 15-745: Introduc9on 1 What Do Compilers Do? 1. Translate one language into another e.g., convert

More information

Introduction. IST557 Data Mining: Techniques and Applications. Jessie Li, Penn State University

Introduction. IST557 Data Mining: Techniques and Applications. Jessie Li, Penn State University Introduction IST557 Data Mining: Techniques and Applications Jessie Li, Penn State University 1 Introduction Why Data Mining? What Is Data Mining? A Mul3-Dimensional View of Data Mining What Kinds of Data

More information

NARCCAP: North American Regional Climate Change Assessment Program. Seth McGinnis, NCAR

NARCCAP: North American Regional Climate Change Assessment Program. Seth McGinnis, NCAR NARCCAP: North American Regional Climate Change Assessment Program Seth McGinnis, NCAR mcginnis@ucar.edu NARCCAP: North American Regional Climate Change Assessment Program Nest highresolution regional

More information

Interna'onal Community for Open and Interoperable AR content and experiences

Interna'onal Community for Open and Interoperable AR content and experiences where professionals get the latest information about standards for AR Interna'onal Community for Open and Interoperable AR content and experiences Summary Report of Fourth Mee'ng Oct 24-25 2011 At- a-

More information

Introduc)on to Matlab

Introduc)on to Matlab Introduc)on to Matlab Marcus Kaiser (based on lecture notes form Vince Adams and Syed Bilal Ul Haq ) MATLAB MATrix LABoratory (started as interac)ve interface to Fortran rou)nes) Powerful, extensible,

More information

Founda'ons of So,ware Engineering. Lecture 11 Intro to QA, Tes2ng Claire Le Goues

Founda'ons of So,ware Engineering. Lecture 11 Intro to QA, Tes2ng Claire Le Goues Founda'ons of So,ware Engineering Lecture 11 Intro to QA, Tes2ng Claire Le Goues 1 Learning goals Define so;ware analysis. Reason about QA ac2vi2es with respect to coverage and coverage/adequacy criteria,

More information

On the Use of Burst Buffers for Accelerating Data-Intensive Scientific Workflows

On the Use of Burst Buffers for Accelerating Data-Intensive Scientific Workflows On the Use of Burst Buffers for Accelerating Data-Intensive Scientific Workflows Rafael Ferreira da Silva, Scott Callaghan, Ewa Deelman 12 th Workflows in Support of Large-Scale Science (WORKS) SuperComputing

More information

Enhancing Feature Interfaces for Suppor8ng So9ware Product Line Maintenance

Enhancing Feature Interfaces for Suppor8ng So9ware Product Line Maintenance Enhancing Feature Interfaces for Suppor8ng So9ware Product Line Maintenance Bruno B. P. Cafeo bcafeo@inf.puc-rio.br LES DI PUC- Rio - Brazil OPUS Group Mo:va:on So9ware Product Line (SPL) Feature Model

More information

Distributed Systems INF Michael Welzl

Distributed Systems INF Michael Welzl Distributed Systems INF 3190 Michael Welzl What is a distributed system (DS)? Many defini8ons [Coulouris & Emmerich] A distributed system consists of hardware and sodware components located in a network

More information

Oracle Applica7on Express (APEX) For E- Business Suite Repor7ng. Your friend in the business.

Oracle Applica7on Express (APEX) For E- Business Suite Repor7ng. Your friend in the business. Oracle Applica7on Express (APEX) For E- Business Suite Repor7ng Your friend in the business. 1 Presenter Jamie Stokes Senior Director Oracle Technology Services Email: jstokes@smartdogservices.com LinkedIn:

More information

The iex.ec Distributed Cloud: Latest Developments and Perspectives. Gilles Fedak

The iex.ec Distributed Cloud: Latest Developments and Perspectives. Gilles Fedak The iex.ec Distributed Cloud: Latest Developments and Perspectives Gilles Fedak (gf@iex.ec) http://iex.ec The Crew Founders Scien2sts Beijing Office Gilles Haiwu Mircea Oleg Ester Cathy Core Developers

More information

Next hop in rou-ng Summary of Future Internet WP1 work. Hannu Flinck

Next hop in rou-ng Summary of Future Internet WP1 work. Hannu Flinck Next hop in rou-ng Summary of Future Internet WP1 work Hannu Flinck Original focus on Rou-ng Scalability Mo$va$on: Internet Architecture Board stated (in RFC 4984): rou-ng scalability is the most important

More information

CCW Workshop Technical Session on Mobile Cloud Compu<ng

CCW Workshop Technical Session on Mobile Cloud Compu<ng CCW Workshop Technical Session on Mobile Cloud Compu

More information

Distributed State Es.ma.on Algorithms for Electric Power Systems

Distributed State Es.ma.on Algorithms for Electric Power Systems Distributed State Es.ma.on Algorithms for Electric Power Systems Ariana Minot, Blue Waters Graduate Fellow Professor Na Li, Professor Yue M. Lu Harvard University, School of Engineering and Applied Sciences

More information

Su#erPatch So.ware Release Notes

Su#erPatch So.ware Release Notes Su#erPatch So.ware Release Notes Version 2.0.0 (build 200); September 1, 2018 New Feature Highlights Free upgrade for all exis1ng users. Su5erPatch 2 comes with Igor Pro version 8. All exis1ng users receive

More information