An Architecture for Large Scale Data Dave Nadeau SDSC Scientific Visualization Group.

Slides:



Advertisements
Similar presentations
COMPUTER GRAPHICS CS 482 – FALL 2014 NOVEMBER 10, 2014 GRAPHICS HARDWARE GRAPHICS PROCESSING UNITS PARALLELISM.
Advertisements

Phillip Dickens, Department of Computer Science, University of Maine. In collaboration with Jeremy Logan, Postdoctoral Research Associate, ORNL. Improving.
VIS Group, University of Stuttgart Tutorial T4: Programmable Graphics Hardware for Interactive Visualization Pre-Integrated Splatting (Stefan Roettger)
Render Cache John Tran CS851 - Interactive Ray Tracing February 5, 2003.
Ray-casting in VolumePro™ 1000
Real-Time Rendering TEXTURING Lecture 02 Marina Gavrilova.
Volume Rendering Volume Modeling Volume Rendering Volume Modeling Volume Rendering 20 Apr
Slide 1 Visualization of scientific data under Linux - Techniques and data wrangling Mike Walterman, Manager of Graphics Programming, Scientific Computing.
Introduction to Volume Rendering Presented by Zvi Devir.
Chapter 6.4 2D Textures and Texture Mapping. 2 Definition and Purpose 2D images designed for use on 3D object Its importance and order in the art pipeline.
High-Quality Video View Interpolation
IN4151 Introduction 3D graphics 1 Introduction to 3D computer graphics part 2 Viewing pipeline Multi-processor implementation GPU architecture GPU algorithms.
Parallel Rendering Ed Angel
Fast Isosurface Visualization on a High-Resolution Scalable Display Wall Adam Finkelstein Allison Klein Kai Li Princeton University Sponsors: DOE, Intel,
The Story So Far The algorithms presented so far exploit: –Sparse sets of images (some data may not be available) –User help with correspondences (time.
Abstract: Digital 3D models are used in industry during the design process. Our client, Siemens PLM, creates software to allow these businesses to view.
Web-based Portal for Discovery, Retrieval and Visualization of Earth Science Datasets in Grid Environment Zhenping (Jane) Liu.
1 Instant replay  The semester was split into roughly four parts. —The 1st quarter covered instruction set architectures—the connection between software.
Hadoop Team: Role of Hadoop in the IDEAL Project ●Jose Cadena ●Chengyuan Wen ●Mengsu Chen CS5604 Spring 2015 Instructor: Dr. Edward Fox.
NATIONAL PARTNERSHIP FOR ADVANCED COMPUTATIONAL INFRASTRUCTURE Discovery Environments Susan L. Graham Chief Computer Scientist Peter.
Developing Reusable Software Infrastructure – Middleware – for Multiscale Modeling Wilfred W. Li, Ph.D. National Biomedical Computation Resource Center.
Leicester, February 24, 2005 VisIVO, a VO-Enabled tool for Scientific Visualization and Data Analysis. VO-TECH Project. Stage01 Ugo Becciani INAF – Astrophysical.
Computer System Architectures Computer System Software
N ATIONAL P ARTNERSHIP FOR A DVANCED C OMPUTATIONAL I NFRASTRUCTURE Brains to Bays --Scaleable Visualization Toolkits Arthur J. Olson Interaction Environments.
Parallel Rendering 1. 2 Introduction In many situations, standard rendering pipeline not sufficient ­Need higher resolution display ­More primitives than.
Visualization Services Group Steve Cutchin – Manager Amit Chourasia – Visualization Scientist Alex DeCastro – Visualization.
National Center for Supercomputing Applications University of Illinois at Urbana–Champaign Practical HPC Visualization Mark Van Moer Visualization Programmer.
A Metadata Based Approach For Supporting Subsetting Queries Over Parallel HDF5 Datasets Vignesh Santhanagopalan Graduate Student Department Of CSE.
1 School of Computer, National University of Defense Technology A Profile on the Grid Data Engine (GridDaEn) Xiao Nong
Adaptive Real-Time Rendering of Planetary Terrains WSCG 2010 Raphaël Lerbour Jean-Eudes Marvie Pascal Gautron THOMSON R&D, Rennes, France.
Edinburgh, January 25, 2005 VisIVO, a VO-Enabled tool for Scientific Visualization and Data Analysis: Overview and Demo 1. Ugo Becciani (OACt): Introduction.
February 2-3, 2006SRB Workshop, San Diego P eter Cao, NCSA Mike Wan, SDSC Sponsored by NLADR, NFS PACI Project in Support of NCSA-SDSC Collaboration Object-level.
The 2000 Decennial Census School District Project: Using Census Data for the School District Mapping System **** Development and Implementation Tai A.
So far we have covered … Basic visualization algorithms Parallel polygon rendering Occlusion culling They all indirectly or directly help understanding.
Light-Weight Data Management Solutions for Scientific Datasets Gagan Agrawal, Yu Su Ohio State Jonathan Woodring, LANL.
ICPP 2012 Indexing and Parallel Query Processing Support for Visualizing Climate Datasets Yu Su*, Gagan Agrawal*, Jonathan Woodring † *The Ohio State University.
Parallel Rendering. 2 Introduction In many situations, a standard rendering pipeline might not be sufficient ­Need higher resolution display ­More primitives.
CS-378: Game Technology Lecture #4: Texture and Other Maps Prof. Okan Arikan University of Texas, Austin V Lecture #4: Texture and Other Maps.
Indexing and Visualizing Multidimensional Data I. Csabai, M. Trencséni, L. Dobos, G. Herczegh, P. Józsa, N. Purger Eötvös University,Budapest.
Tone Mapping on GPUs Cliff Woolley University of Virginia Slides courtesy Nolan Goodnight.
Media Arts and Technology Graduate Program UC Santa Barbara MAT 259 Visualizing Information Winter 2006George Legrady1 MAT 259 Visualizing Information.
 proposed work This project aims to design and develop a framework for terrain visualization flexible enough to allow arbitrary visualization of terrain.
Major Disciplines in Computer Science Ken Nguyen Department of Information Technology Clayton State University.
GVis: Grid-enabled Interactive Visualization State Key Laboratory. of CAD&CG Zhejiang University, Hangzhou
CS559: Computer Graphics Lecture 4: Compositing and Resampling Li Zhang Spring 2008.
Data Management for Decision Support Session-4 Prof. Bharat Bhasker.
Supported By Understanding the dynamics of the hydrological phenomena associated to wetlands requires analyzing data gathered from.
Center for Computational Visualization University of Texas, Austin Visualization and Graphics Research Group University of California, Davis Molecular.
SDSC The Scripps Research Institute U. Texas, Austin U.C.L.A. SIO/MIT U. California, Davis Mississippi State U. Scalable Visualization Toolkits for Bays.
Biomedical Informatics Research Network Pipelines & Processing: Tools & Toolkits David Rex, John Moreland October 9, nd Annual All Hands Meeting.
Client-Server Paradise ICOM 8015 Distributed Databases.
Visualization Four groups Design pattern for information visualization
CHAPTER 4 THE VISUALIZATION PIPELINE. CONTENTS The focus is on presenting the structure of a complete visualization application, both from a conceptual.
Partnerships in Innovation: Serving a Networked Nation Grid Technologies: Foundations for Preservation Environments Portals for managing user interactions.
Tackling I/O Issues 1 David Race 16 March 2010.
67 x 89 = ? 67 x
Creation and Visualization of 3D Scenes with the MRPT library January, 2007 Jose Luis Blanco Claraco Dept. of Automation and System Engineering University.
1 Copyright © 2008, Oracle. All rights reserved. Repository Basics.
IMAGE PROCESSING is the use of computer algorithms to perform image process on digital images   It is used for filtering the image and editing the digital.
COMPUTER GRAPHICS CHAPTER 38 CS 482 – Fall 2017 GRAPHICS HARDWARE
So far we have covered … Basic visualization algorithms
Self-Organizing Maps for Content-Based Image Database Retrieval
Introduction to Computer Graphics with WebGL
Static Image Filtering on Commodity Graphics Processors
Volume Rendering Lecture 21.
Declarative Transfer Learning from Deep CNNs at Scale
CS-378: Game Technology Lecture #4: Texture and Other Maps
Database System Architectures
Point Cloud Processing
L. Glimcher, R. Jin, G. Agrawal Presented by: Leo Glimcher
Presentation transcript:

An Architecture for Large Scale Data Dave Nadeau SDSC Scientific Visualization Group

Motivation CT Cryosection Classification Support analysis, filtering, and compositingSupport analysis, filtering, and compositing –Larger-than-core (and swap) data sets –Multi-modal and time-varying data –Multiple data sets simultaneously And...And... –Do efficient data movement –Execute well on parallel architectures –Integrate easily w/existing applications & toolkits Support Alpha project applicationsSupport Alpha project applications

Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit Layered Toolkit Architecture Manage an N-space data grid Cache pages for lazy I/O Support specific file formats Manage file storage Bind a coord. system to data Orchestrate filter execution

Managing Data Grids Manage a paged data grid (array-like)Manage a paged data grid (array-like) –An N-dimensional grid of cells –Spatial data & time-series –Arbitrary cell data content Handle larger-than-core dataHandle larger-than-core data –Transparently pages data in/out –Support from ADR & DataCutter –Compressed data (disk & memory) Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit Data Grid Toolkit

Random access (slow)Random access (slow) –Get/set cells in any order Structured access (faster)Structured access (faster) –Get/set cells in a pre-defined order Data-order access (fastest)Data-order access (fastest) –Get/set cells in the data’s storage order Pre-fetching Intelligently Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit Data Grid Toolkit

Paging Intelligently Neighborhood-aware pagingNeighborhood-aware paging –Page in nearby cells in N dimensions –Support convolution filtering, rendering, marching-cubes, Current center cell Keep neighboring cells paged-in as well Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit Data Grid Toolkit Filter window

Using Coordinate Systems Bind a coordinate system to a data gridBind a coordinate system to a data grid –Euclidean, cylindrical, spherical, time-series,... –Uniform, structured, unstructured Handle coordinate system-based operationsHandle coordinate system-based operations –Resampling with interpolation –Lazy-evaluation Multiple file format handlersMultiple file format handlers Application Data Grid Toolkit Data Management File Format Handling SRB,ADR, etc. Mesh Toolkit Expression Tree Toolkit Mesh Toolkit

Operating on Data Define an expression tree for data operationsDefine an expression tree for data operations –Leaf nodes are data sets, functions,... –Interior nodes are composite, filter,... –Transforms align overlapping data sets Execute it to generate samplesExecute it to generate samples –Client defines the expression –Server on big iron executes it Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit Client Server

Operating on Expressions Expressions can be optimizedExpressions can be optimized –Re-order operators –Similar to optimizing compilers & databases Sample order can be optimizedSample order can be optimized –Re-order data accesses for better cache efficiency Data can be staged & intermediate results cachedData can be staged & intermediate results cached Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit

Combining Brain Data Sets RGB to HSI Scalar to RGB Mask by Hue Scalar CT-scan Color Cryosection Color Segmentation Extract Hue Composite 512 x 512 x x 710 x 672 Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit

Combining Brain Data Sets CT Cryosection Composited Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit

Combining Stellar Data Sets Complex expression treesComplex expression trees –60+ nodes in the Orion body 90+ separate expression trees90+ separate expression trees –Orion, proplyds, shock fronts,... Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit

And more toolkits... Interactive imaging with...Interactive imaging with... –Mitsubishi VolumePro cards –Point clouds & 3D texture mapping with graphics pipelines High-quality imaging with VISTA...High-quality imaging with VISTA... Application Data Grid Toolkit Data Management File Format Handling SRB, ADR, etc. Mesh Toolkit Expression Tree Toolkit Other Toolkits VolumePro Point Cloud VISTA 3D Texture

Design Team Scripps Research Art Olson Mike Pique Michel Sanner SDSC Bernard Pailthorpe Dave Nadeau Jon Genetti John Moreland Mike Bailey Rich Charles Alex Decastro U. Texas Chandrajit Bajaj Ariel Shamir

Data-Visualization Pipeline Get data from disk efficiently Manage data in memory efficiently Compute on data efficiently Visualize data efficiently ComputationVisualization Data SRB Server MCAT (Metadata) ADR DataCutter SRB Server KeLP FloorPlan Data Data Orchestration...

Data-Visualization Pipeline Get data from disk efficiently Manage data in memory efficiently Compute on data efficiently Visualize data efficiently ComputationVisualization Data Data - Vis Toolkits Interaction Tools VISTA Renderer Data Orchestration...