xxviii Contents
12 Hybrid Parallelism 261
E. Wes Bethel, David Camp, Hank Childs, Christoph Garth, Mark
Howison, Kenneth I. Joy, and David Pugmire
12.1Introduction ........................... 262
12.2HybridParallelismandVolumeRendering .......... 264
12.2.1 BackgroundandPreviousWork............. 264
12.2.2 Implementation...................... 265
12.2.2.1 Shared-MemoryParallelRayCasting .... 266
12.2.2.2 ParallelCompositing ............. 266
12.2.3 ExperimentMethodology ................ 267
12.2.4 Results .......................... 268
12.2.4.1 Initialization .................. 268
12.2.4.2 Ghost Data/Halo Exchange .......... 269
12.2.4.3 RayCasting .................. 269
12.2.4.4 Compositing .................. 272
12.2.4.5 OverallPerformance.............. 272
12.3HybridParallelismandIntegralCurveCalculation ...... 275
12.3.1 BackgroundandContext ................ 275
12.3.2 DesignandImplementation............... 276
12.3.2.1 ParallelizeOverSeeds............. 276
12.3.2.2 ParallelizeOverBlocks ............ 277
12.3.3 ExperimentMethodology ................ 278
12.3.3.1 Factors Influencing Parallelization Strategy . 278
12.3.3.2 TestCases ................... 279
12.3.3.3 RuntimeEnvironment............. 279
12.3.3.4 Measurements ................. 280
12.3.4 Results .......................... 280
12.3.4.1 ParallelizationOverSeeds........... 280
12.3.4.2 ParallelizationOverBlocks.......... 282
12.4ConclusionandFutureWork .................. 283
13 Visualization at Extreme Scale Concurrency 291
Hank Childs, David Pugmire, Sean Ahern, Brad Whitlock, Mark
Howison, Prabhat, Gunther Weber, and E. Wes Bethel
13.1Overview—PureParallelism .................. 292
13.2MassiveDataExperiments ................... 293
13.2.1 VaryingoverSupercomputingEnvironment ...... 296
13.2.2 VaryingoverI/OPattern ................ 297
13.2.3 VaryingoverDataGeneration ............. 298
13.3ScalingExperiments ....................... 299
13.3.1 StudyOverview ..................... 299
13.3.2 Results .......................... 300
13.4PitfallsatScale ......................... 301
13.4.1 VolumeRendering .................... 301
13.4.2 All-to-OneCommunication ............... 303
Contents xxix
13.4.3 SharedLibrariesandStart-upTime .......... 304
13.5Conclusion ............................ 305
14 Performance Optimization and Auto-Tuning 307
E. Wes Bethel and Mark Howison
14.1Introduction ........................... 308
14.2 Optimizing Performance of a 3D Stencil Operator on the GPU 310
14.2.1 IntroductionandRelatedWork............. 310
14.2.2 DesignandMethodology ................ 312
14.2.3 Results .......................... 313
14.2.3.1 Algorithmic Design Option: Width-, Height-,
andDepth-RowKernels............ 313
14.2.3.2 Device-Specific Feature: Constant Versus Global
Memory for Filter Weights .......... 314
14.2.3.3 Tunable Algorithmic Parameter: Thread Block
Size ....................... 314
14.2.4 LessonsLearned ..................... 317
14.3 Optimizing Ray Casting Volume Rendering on Multi-Core
GPUsandMany-CoreGPUs .................. 317
14.3.1 IntroductionandRelatedWork............. 317
14.3.2 DesignandMethodology ................ 319
14.3.3 Results .......................... 320
14.3.3.1 Tunable Parameter: Image Tile Size/CUDA
BlockSize ................... 320
14.3.3.2 Algorithmic Optimization: Early Ray Termi-
nation...................... 323
14.3.3.3 Algorithmic Optimization: Z-Ordered Mem-
ory ....................... 324
14.3.4 LessonsLearned ..................... 325
14.4Conclusion ............................ 326
15 The Path to Exascale 331
Sean Ahern
15.1Introduction ........................... 332
15.2FutureSystemArchitectures .................. 332
15.3ScienceUnderstandingNeedsattheExascale ......... 335
15.4 Research Directions ....................... 338
15.4.1 Data Processing Modes ................. 338
15.4.1.1 In Situ Processing . .............. 338
15.4.1.2 Post-Processing Data Analysis ........ 339
15.4.2 VisualizationandAnalysisMethods .......... 341
15.4.2.1 Support for Data Processing Modes ..... 341
15.4.2.2 TopologicalMethods.............. 342
15.4.2.3 StatisticalMethods .............. 343
15.4.2.4 Adapting to Increased Data Complexity . . . 343
xxx Contents
15.4.3 I/OandStorageSystems ................ 344
15.4.3.1 StorageTechnologiesfortheExascale .... 345
15.4.3.2 I/OMiddlewarePlatforms .......... 346
15.5ConclusionandthePathForward ............... 347
IV High Performance Visualization Implementa-
tions 355
16 VisIt: An End-User Tool for Visualizing and Analyzing Very
Large Data 357
Hank Childs, Eric Brugger, Brad Whitlock, Jeremy Meredith, Sean
Ahern, David Pugmire, Kathleen Biagas, Mark Miller, Cyrus
Harrison, Gunther H. Weber, Hari Krishnan, Thomas Fogal, Allen
Sanderson, Christoph Garth, E. Wes Bethel, David Camp, Oliver
R¨ubel, and Marc Durant, Jean M. Favre, Paul Navr´atil
16.1Introduction ........................... 358
16.2FocalPoints ........................... 359
16.2.1 EnableDataUnderstanding............... 359
16.2.2 Support for Large Data ................. 360
16.2.3 Provide a Robust and Usable Product for End Users . 360
16.3Design .............................. 361
16.3.1 Architecture ....................... 361
16.3.2 Parallelism ........................ 362
16.3.3 User Interface Concepts and Extensibility ....... 363
16.3.4 TheSizeandBreadthofVisIt ............. 364
16.4 Successes ............................. 364
16.4.1 Scalability Successes ................... 365
16.4.2 ARepositoryforLargeDataAlgorithms........ 365
16.4.3 SupercomputingResearchPerformedwithVisIt.... 366
16.4.4 User Successes ...................... 366
16.5FutureChallenges ........................ 368
16.6Conclusion ............................ 368
17 IceT 373
Kenneth Moreland
17.1Introduction ........................... 373
17.2Motivation ............................ 374
17.3Implementation ......................... 374
17.3.1 Theoretical Limitations ... and How to Break Them . 375
17.3.2 PixelReductionTechniques............... 376
17.3.3 TrickstoBoosttheFrameRate............. 377
17.4ApplicationProgrammingInterface .............. 378
17.4.1 ImageGeneration .................... 378
17.4.2 OpaqueversusTransparentRendering......... 379
17.5Conclusion ............................ 379
Contents xxxi
18 The ParaView Visualization Application 383
Utkarsh Ayachit, Berk Geveci, Kenneth Moreland, and John Patchett,
Jim Ahrens
18.1Introduction ........................... 384
18.2UnderstandingtheNeed .................... 384
18.3TheParaViewFramework ................... 386
18.3.1 Configurations ...................... 387
18.4 Parallel Data Processing .................... 387
18.5TheParaViewApplication ................... 390
18.5.1 GraphicalUserInterface................. 390
18.5.2 ScriptingwithPython.................. 391
18.6CustomizingwithPlug-insandCustomApplications .... 391
18.7 Co-Processing: In Situ Visualization and Data Analysis . . . 392
18.8ParaViewWeb:InteractiveVisualizationfortheWeb ..... 393
18.9ParaViewInUse ......................... 394
18.9.1 Identifying and Validating Fragmentation in Shock
PhysicsSimulation.................... 394
18.9.2 ParaView at the Los Alamos National Laboratory . . 396
18.9.3 Analyzing Simulations of the Earth’s Magnetosphere . 397
18.10Conclusion ............................ 398
19 The ViSUS Visualization Framework 401
Valerio Pascucci, Giorgio Scorzelli, Brian Summa, Peer-Timo Bremer,
Attila Gyulassy, Cameron Christensen, Sujin Philip, and Sidharth Kumar
19.1Introduction ........................... 402
19.2ViSUSSoftwareArchitecture .................. 402
19.3Applications ........................... 408
20 The VAPOR Visualization Application 415
Alan Norton and John Clyne
20.1Introduction ........................... 415
20.1.1 Features.......................... 416
20.1.2 Limitations ........................ 417
20.2 Progressive Data Access ..................... 417
20.2.1 VAPORDataCollection................. 418
20.2.2 Multiresolution...................... 419
20.3Visualization-GuidedAnalysis ................. 420
20.4 Progressive Access Examination ................ 422
20.4.1 Discussion......................... 423
20.5Conclusion ............................ 424
21 The EnSight Visualization Application 429
Randall Frank and Michael F. Krogh
21.1Introduction ........................... 429
21.2 EnSight Architectural Overview ................ 430
xxxii Contents
21.3ClusterAbstraction:CEIShell ................. 432
21.3.1 VirtualClusteringViaCEIShellRoles ......... 433
21.3.2 ApplicationInvocation.................. 434
21.3.3 CEIShell Extensibility .................. 434
21.4AdvancedRendering....................... 435
21.4.1 CustomizedFragmentRendering ............ 435
21.4.2 ImageCompositionSystem ............... 438
21.5Conclusion ............................ 440
Index 443
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.