Scalability of Uintah Past Present and Future

Size: px
Start display at page:

Download "Scalability of Uintah Past Present and Future"

Transcription

1 DOE for funding the CSAFE project (97-10), DOE NETL, DOE NNSA NSF for funding via SDCI and PetaApps, INCITE, XSEDE Scalability of Uintah Past Present and Future Martin Berzins Qingyu Meng John Schmidt, Alan Humphrey, Justin Luitjens How do we get software frameworks for complex multiscale multiphysics problems such as Uintah to scale on machines such as Blue Waters, Sequoia, Titan and the next two generations of machines? Directed Acyclic Graph approaches are seen as a key technology. Exascale programming will require prioritization of critical-path and noncritical path tasks, adaptive directed acyclic graph scheduling of critical- Exascale Report

2 Directed Acyclic Graph Based Applications/Languages 1: 1 1: 2 1: 3 1: 4 2: 2 2: 3 2: 4 Charm++: Object-based Virtualization NAMD etc etc Intel CnC: new language for graph based parallelism 2: 2 Plasma (Dongarra): DAG based Parallel linear algebra software 2: 3 3: 3 3: 3 2: 4 3: 4 Uintah Software

3 Uintah Applications Explosions Angiogenesis Micropin Flow Industrial Flares Explosions Sandstone Compaction Foam Compaction Carbon capture and cleanup Shaped Charges

4 ICE is a cell-centered finite volume method for Navier Stokes equations Uintah Patch and Variables Structured Grid Variable (for Flows) Cell Centered, Node Centered,Face Centered Unstructured Points (for Solids) Particles ARCHES is a combustion code using several different radiation models and linear solvers MPM is a novel method that uses particles and nodes Exchange data with ICE, not just boundary condition

5 Utah Uintah Software Parallelism AMR Meshes Structured Grid + Unstructured Points Patch-based Domain Decomposition Adaptive Mesh Refinement Dynamic Load Balancing Profiling + Forecasting Model Parallel Space Filling Curves Works on MPI and/or thread level Uses asynchronous task directed graph approach to specify the relationships between tasks not when they will compute. Task Graph for Material Point Method

6 Uintah uses a combination of compile-time and run-time techniques xml Task Compile (when needed) Run Time (each timestep) Parallel I/O

7 Uintah Task Graph 4 patches single level ICE task graph Compiled from detailed tasks (on local and neighboring patches) Intermediate Representation for Uintah runtime system Internal dependencies ->Edges External dependencies ->MPI message tags Variable dictionary held in data warehouse Stored locally not globally

8 Scalability of Uintah DAG Framework This is what scalability with a large software framework such as Uintah feels like The Harry Potter approach to scalability??? You only need a few very smart young (computer science know what impossible means. Steve Parker designer of Uintah Justin Luitjens Qingyu Meng Alan Humphrey plus a strong supporting cast.

9 The DAG Approach is not a silver bullet Uintah Phase overlap communications with computation. Static task graph execution standard data structures one MPI process per core. No AMR. Uintah Phase improved fast data structures, load balancer. AMR to 12k cores, then 100K cores using new approach before data structures cause problems. Out of order and dynamic task execution. Uintah Phase Hybrid model. Theaded runtine system on node. One MPI process and one data warehouse per node. Multiple cores and GPUs grab tasks as needed. Fast scalable use of hypre for linear equations.

10 Weak and Strong Scalability: Problem size n on p cores takes time T(n,p) Strong Scalability Tnp (, ) Tn (,1)/ p Try to solve the same problem p times more quickly on p cores Weak Scalability Tnpp (, ) Tn (,1) Solve a problem that is p times as large in the same time on p cores Theorem Both weak and strong scalability only if linear complexity [Tirado + Martin] 1998 Tn (,1) n

11 Scalable Tiled AMR Regridding Algorithm [Analysis in Concurrency] Tiled Algorithm Tiles that contain flags become patches Simple and easy to parallelize Semi-regular patch sets that can be exploited Example: Neighbor finding Each core processes subset of refinement flags in parallel-helps produce global patch set GBRv1 vs Tiled

12 Strong Scaling of AMR Algorithms problem size fixed as number of cores increases should lead to decreasing execution time. GBR is Global Berger-Rigoutsos No strong scaling Strong Scaling T: GBRv1 = c1 (Flags/Proc) log(npatches ) + c2 M(GBRv1) T: Tiled = c5 Nmeshcells / Proc T: LBR = GBR on a compute node Dots are data and lines are models

13 Example: AMR MPMICE A PBX explosive flow quickly pushing a piece of its metal container Click Flow velocity and particle volume Computational grids and particles Grid Variables: Fixed number per patch, relative easy to balance Particle Variables: Variable number per patch, hard to load balance

14 Uintah Scalability(Original MPI-only) Poor scalability up to 98K cores (Kraken, NICS) Issues: Two flows transporting One flow with particles moving Out of memory with 98K cores AMR MPMICE scaling, Load Imbalance Solution: New runtime system with Hybrid thread/mpi

15 One Data Warehouse per mode Hybrid threaded CPU GPU scheduler No on node MPI Better load balance Per node (more tasks) Lower MPI reduction Time. Atomic operations for Faster lock-free memory access Needs thread-safe code Originally one control core now decentralized -GPUs and CPUs pull work

16 New Hybrid Model Memory Savings: Ghost Cells MPI: Thread/MPI: Local Patch Ghost Cells Raw Data: doubles doubles MPI buffer: doubles doubles Total: 75K doubles 40K doubles (example on Kraken, 12 cores/node, more cores more savings)

17 Memory Savings Global Meta-data copies 60 bytes or 7.5 doubles per patch Each copy per core vs Each copy per node MPI library buffer overhead Results: cores percent 61% 47% 36% 27% 18% 11% AMRICE: Simulation of the transport of two fluids with a prescribed initial velocity of Mach two: 435 million cells, strong scaling runs on Kraken

18 MPI Patch Boundary Thread/MPI Processor Boundary Particle Reallocations Move particles to new patches After each timestep Cross patch boundary: just re-indexing Cross processor boundary: MPI scatter record (expensive!) Thread/MPI Fewer particles cross processor boundary

19 Results: Scalability on Jaguar One flow with particles moving 3-level AMR MPM ICE Two flows transporting, 3-level AMR ICE Using one control core per node scales up to 196K cores

20 OLD Scalability on Titan Scaling Breakdown One flow with particles moving 3-level AMR MPM ICE 50% efficiency At 256K cores vs 16K cores Distributed Controller

21 ICES Carbon Dioxide Cleanup Red is C02 Turbulent flow problem - need to quantify the uncertainty in the Simulation to estimate how much CO2 is removed. Need at least 100K cores to resolve the problem scales.

22 4 months work with hypre PFMG preconditioner and CG solver. Hypre setup only called once and workspace carried by Uintah Weak Scalability with hypre on turbulent combustion problems Uintah on implicit problems using hypre solver Good approx log(c) ( weak scaling to 98K cores on turbulent combustion Helium Plume problem with Red-Black Gauss Seidel on Kraken XT5. Jacobi iteration on Jaguar XK6 has weak scaling, c = cores Kraken case has 2x Jaguar unknowns per core

23 NVIDIA 2050/2070 vs Intel xeon 2x4 cores DESIGNING FOR EXASCALE Clear trend towards accelerators e.g. GPU but also Intel MIC new NSF -. 15PF Balance factor = flops/bandwidth - high -based codes,2x over multicore cpu estimated and achieved for ICE. Similar results by others. Network and memory performance more slowly growing than cpu/gpu performance GPU performance of ray-tracing radiation method is 100x cpu See his talk this week Overlapping and hiding Communications essential new MPI non-blocking collective will help- new algorithms needed.

24 An Exascale Problem Design of Alstom Clean coal Boilers Temperature field For 350MWe boiler problem. LES resolution needed: 1mm per side for each computational volume = 9x 1012 cells Based on initial runs - to run in 48 hours of wall clock time requires M fast cores. Professor Phil Smith ICSE Utah

25 Summary Scalability still a challenge even with DAG approach which does work amazingly well Uintah Hybrid Parallelism Domain Decomposition + Task Graph MPI Level load balance based on patches Multi-thread task graph execution locally Concurrent data warehouse Thread task queue controller Reduce overhead of global metadata and ghost cells Memory savings: 40%~90% Scaling No in-node MPI, better load balancing AMR MPMICE 2~4X faster when running with 12k+ cores Allow Uintah scaling to 256K cores even for fluid-structure calculations GPU development ongoing The approach used here shows promise for very large core and GPU counts NSF funded work will stop in September 2013

Preliminary Experiences with the Uintah Framework on on Intel Xeon Phi and Stampede

Preliminary Experiences with the Uintah Framework on on Intel Xeon Phi and Stampede Preliminary Experiences with the Uintah Framework on on Intel Xeon Phi and Stampede Qingyu Meng, Alan Humphrey, John Schmidt, Martin Berzins Thanks to: TACC Team for early access to Stampede J. Davison

More information

The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System

The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System Alan Humphrey, Qingyu Meng, Martin Berzins Scientific Computing and Imaging Institute & University of Utah I. Uintah Overview

More information

Radiation Modeling Using the Uintah Heterogeneous CPU/GPU Runtime System

Radiation Modeling Using the Uintah Heterogeneous CPU/GPU Runtime System Radiation Modeling Using the Uintah Heterogeneous CPU/GPU Runtime System Alan Humphrey, Qingyu Meng, Martin Berzins, Todd Harman Scientific Computing and Imaging Institute & University of Utah I. Uintah

More information

Alan Humphrey, Qingyu Meng, Brad Peterson, Martin Berzins Scientific Computing and Imaging Institute, University of Utah

Alan Humphrey, Qingyu Meng, Brad Peterson, Martin Berzins Scientific Computing and Imaging Institute, University of Utah Alan Humphrey, Qingyu Meng, Brad Peterson, Martin Berzins Scientific Computing and Imaging Institute, University of Utah I. Uintah Framework Overview II. Extending Uintah to Leverage GPUs III. Target Application

More information

Cost Estimation Algorithms for Dynamic Load Balancing of AMR Simulations

Cost Estimation Algorithms for Dynamic Load Balancing of AMR Simulations Cost Estimation Algorithms for Dynamic Load Balancing of AMR Simulations Justin Luitjens, Qingyu Meng, Martin Berzins, John Schmidt, et al. Thanks to DOE for funding since 1997, NSF since 2008, TACC, NICS

More information

A Unified Approach to Heterogeneous Architectures Using the Uintah Framework

A Unified Approach to Heterogeneous Architectures Using the Uintah Framework DOE for funding the CSAFE project (97-10), DOE NETL, DOE NNSA NSF for funding via SDCI and PetaApps A Unified Approach to Heterogeneous Architectures Using the Uintah Framework Qingyu Meng, Alan Humphrey

More information

An Applications-Driven path from Terascale to Exascale (?) with the Asynchronous Many Task (AMT) Uintah Framework

An Applications-Driven path from Terascale to Exascale (?) with the Asynchronous Many Task (AMT) Uintah Framework An Applications-Driven path from Terascale to Exascale (?) with the Asynchronous Many Task (AMT) Uintah Framework Martin Berzins Scientific Imaging and Computing Institute University of Utah 1. Foundations:

More information

Improving Uintah s Scalability Through the Use of Portable

Improving Uintah s Scalability Through the Use of Portable Improving Uintah s Scalability Through the Use of Portable Kokkos-Based Data Parallel Tasks John Holmen1, Alan Humphrey1, Daniel Sunderland2, Martin Berzins1 University of Utah1 Sandia National Laboratories2

More information

Petascale Parallel Computing and Beyond. General trends and lessons. Martin Berzins

Petascale Parallel Computing and Beyond. General trends and lessons. Martin Berzins Petascale Parallel Computing and Beyond General trends and lessons Martin Berzins 1. Technology Trends 2. Towards Exascale 3. Trends in programming large scale systems What kind of machine will you use

More information

The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System

The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System The Uintah Framework: A Unified Heterogeneous Task Scheduling and Runtime System Qingyu Meng Scientific Computing and Imaging Institute University of Utah Salt Lake City, UT 84112 USA Email: qymeng@sci.utah.edu

More information

Solving Petascale Turbulent Combustion Problems with the Uintah Software

Solving Petascale Turbulent Combustion Problems with the Uintah Software Solving Petascale Turbulent Combustion Problems with the Uintah Software Martin Berzins DOE NNSA PSAAP2 Center Thanks to DOE ASCI (97-10), NSF, DOE NETL+NNSA, NSF, INCITE, XSEDE, ALCC, ORNL, ALCF for funding

More information

Past, Present and Future Scalability of the Uintah Software

Past, Present and Future Scalability of the Uintah Software Past, Present and Future Scalability of the Uintah Software Martin Berzins mb@cs.utah.edu John Schmidt John.Schmidt@utah.edu Alan Humphrey ahumphre@cs.utah.edu Qingyu Meng qymeng@cs.utah.edu ABSTRACT The

More information

Preliminary Experiences with the Uintah Framework on Intel Xeon Phi and Stampede

Preliminary Experiences with the Uintah Framework on Intel Xeon Phi and Stampede 1 Preliminary Experiences with the Uintah Framework on Intel Xeon Phi and Stampede Qingyu Meng, Alan Humphrey, John Schmidt, Martin Berzins UUSCI-2013-002 Scientific Computing and Imaging Institute March

More information

Automatic Halo Management for the Uintah GPU-Heterogeneous Asynchronous Many-Task Runtime

Automatic Halo Management for the Uintah GPU-Heterogeneous Asynchronous Many-Task Runtime 1 Automatic Halo Management for the Uintah GPU-Heterogeneous Asynchronous Many-Task Runtime Brad Peterson, Alan Humphrey, Dan Sunderland, James Sutherland, Tony Saad, Harish Dasari, Martin Berzins Scientific

More information

Investigating Applications Portability with the Uintah DAG-based Runtime System on PetaScale Supercomputers

Investigating Applications Portability with the Uintah DAG-based Runtime System on PetaScale Supercomputers Investigating Applications Portability with the Uintah DAG-based Runtime System on PetaScale Supercomputers Qingyu eng, Alan Humphrey, John Schmidt, artin Berzins UUSCI-- Scientific Computing and Imaging

More information

Introducing Overdecomposition to Existing Applications: PlasComCM and AMPI

Introducing Overdecomposition to Existing Applications: PlasComCM and AMPI Introducing Overdecomposition to Existing Applications: PlasComCM and AMPI Sam White Parallel Programming Lab UIUC 1 Introduction How to enable Overdecomposition, Asynchrony, and Migratability in existing

More information

CMSC 714 Lecture 6 MPI vs. OpenMP and OpenACC. Guest Lecturer: Sukhyun Song (original slides by Alan Sussman)

CMSC 714 Lecture 6 MPI vs. OpenMP and OpenACC. Guest Lecturer: Sukhyun Song (original slides by Alan Sussman) CMSC 714 Lecture 6 MPI vs. OpenMP and OpenACC Guest Lecturer: Sukhyun Song (original slides by Alan Sussman) Parallel Programming with Message Passing and Directives 2 MPI + OpenMP Some applications can

More information

HARNESSING IRREGULAR PARALLELISM: A CASE STUDY ON UNSTRUCTURED MESHES. Cliff Woolley, NVIDIA

HARNESSING IRREGULAR PARALLELISM: A CASE STUDY ON UNSTRUCTURED MESHES. Cliff Woolley, NVIDIA HARNESSING IRREGULAR PARALLELISM: A CASE STUDY ON UNSTRUCTURED MESHES Cliff Woolley, NVIDIA PREFACE This talk presents a case study of extracting parallelism in the UMT2013 benchmark for 3D unstructured-mesh

More information

Forest-of-octrees AMR: algorithms and interfaces

Forest-of-octrees AMR: algorithms and interfaces Forest-of-octrees AMR: algorithms and interfaces Carsten Burstedde joint work with Omar Ghattas, Tobin Isaac, Georg Stadler, Lucas C. Wilcox Institut für Numerische Simulation (INS) Rheinische Friedrich-Wilhelms-Universität

More information

Enabling In Situ Viz and Data Analysis with Provenance in libmesh

Enabling In Situ Viz and Data Analysis with Provenance in libmesh Enabling In Situ Viz and Data Analysis with Provenance in libmesh Vítor Silva Jose J. Camata Marta Mattoso Alvaro L. G. A. Coutinho (Federal university Of Rio de Janeiro/Brazil) Patrick Valduriez (INRIA/France)

More information

Generic finite element capabilities for forest-of-octrees AMR

Generic finite element capabilities for forest-of-octrees AMR Generic finite element capabilities for forest-of-octrees AMR Carsten Burstedde joint work with Omar Ghattas, Tobin Isaac Institut für Numerische Simulation (INS) Rheinische Friedrich-Wilhelms-Universität

More information

Experiences with ENZO on the Intel Many Integrated Core Architecture

Experiences with ENZO on the Intel Many Integrated Core Architecture Experiences with ENZO on the Intel Many Integrated Core Architecture Dr. Robert Harkness National Institute for Computational Sciences April 10th, 2012 Overview ENZO applications at petascale ENZO and

More information

Software and Performance Engineering for numerical codes on GPU clusters

Software and Performance Engineering for numerical codes on GPU clusters Software and Performance Engineering for numerical codes on GPU clusters H. Köstler International Workshop of GPU Solutions to Multiscale Problems in Science and Engineering Harbin, China 28.7.2010 2 3

More information

Introduction to parallel Computing

Introduction to parallel Computing Introduction to parallel Computing VI-SEEM Training Paschalis Paschalis Korosoglou Korosoglou (pkoro@.gr) (pkoro@.gr) Outline Serial vs Parallel programming Hardware trends Why HPC matters HPC Concepts

More information

Large Scale Parallel Solution of Incompressible Flow Problems using Uintah and hypre

Large Scale Parallel Solution of Incompressible Flow Problems using Uintah and hypre 1 Large Scale Parallel Solution of Incompressible Flow Problems using Uintah and hypre John Schmidt, Martin Berzins, Jeremy Thornock, Tony Saad, James Sutherland UUSCI-2012-002 Scientific Computing and

More information

Optimisation of LESsCOAL for largescale high-fidelity simulation of coal pyrolysis and combustion

Optimisation of LESsCOAL for largescale high-fidelity simulation of coal pyrolysis and combustion Optimisation of LESsCOAL for largescale high-fidelity simulation of coal pyrolysis and combustion Kaidi Wan 1, Jun Xia 2, Neelofer Banglawala 3, Zhihua Wang 1, Kefa Cen 1 1. Zhejiang University, Hangzhou,

More information

EXTENDING THE UINTAH FRAMEWORK THROUGH THE PETASCALE MODELING OF DETONATION IN ARRAYS OF HIGH EXPLOSIVE DEVICES

EXTENDING THE UINTAH FRAMEWORK THROUGH THE PETASCALE MODELING OF DETONATION IN ARRAYS OF HIGH EXPLOSIVE DEVICES EXTENDING THE UINTAH FRAMEWORK THROUGH THE PETASCALE MODELING OF DETONATION IN ARRAYS OF HIGH EXPLOSIVE DEVICES MARTIN BERZINS, JACQUELINE BECKVERMIT, TODD HARMAN, ANDREW BEZDJIAN, ALAN HUMPHREY, QINGYU

More information

Massively Parallel Finite Element Simulations with deal.ii

Massively Parallel Finite Element Simulations with deal.ii Massively Parallel Finite Element Simulations with deal.ii Timo Heister, Texas A&M University 2012-02-16 SIAM PP2012 joint work with: Wolfgang Bangerth, Carsten Burstedde, Thomas Geenen, Martin Kronbichler

More information

Enzo-P / Cello. Scalable Adaptive Mesh Refinement for Astrophysics and Cosmology. San Diego Supercomputer Center. Department of Physics and Astronomy

Enzo-P / Cello. Scalable Adaptive Mesh Refinement for Astrophysics and Cosmology. San Diego Supercomputer Center. Department of Physics and Astronomy Enzo-P / Cello Scalable Adaptive Mesh Refinement for Astrophysics and Cosmology James Bordner 1 Michael L. Norman 1 Brian O Shea 2 1 University of California, San Diego San Diego Supercomputer Center 2

More information

Efficient AMG on Hybrid GPU Clusters. ScicomP Jiri Kraus, Malte Förster, Thomas Brandes, Thomas Soddemann. Fraunhofer SCAI

Efficient AMG on Hybrid GPU Clusters. ScicomP Jiri Kraus, Malte Förster, Thomas Brandes, Thomas Soddemann. Fraunhofer SCAI Efficient AMG on Hybrid GPU Clusters ScicomP 2012 Jiri Kraus, Malte Förster, Thomas Brandes, Thomas Soddemann Fraunhofer SCAI Illustration: Darin McInnis Motivation Sparse iterative solvers benefit from

More information

General Plasma Physics

General Plasma Physics Present and Future Computational Requirements General Plasma Physics Center for Integrated Computation and Analysis of Reconnection and Turbulence () Kai Germaschewski, Homa Karimabadi Amitava Bhattacharjee,

More information

Two-Phase flows on massively parallel multi-gpu clusters

Two-Phase flows on massively parallel multi-gpu clusters Two-Phase flows on massively parallel multi-gpu clusters Peter Zaspel Michael Griebel Institute for Numerical Simulation Rheinische Friedrich-Wilhelms-Universität Bonn Workshop Programming of Heterogeneous

More information

14MMFD-34 Parallel Efficiency and Algorithmic Optimality in Reservoir Simulation on GPUs

14MMFD-34 Parallel Efficiency and Algorithmic Optimality in Reservoir Simulation on GPUs 14MMFD-34 Parallel Efficiency and Algorithmic Optimality in Reservoir Simulation on GPUs K. Esler, D. Dembeck, K. Mukundakrishnan, V. Natoli, J. Shumway and Y. Zhang Stone Ridge Technology, Bel Air, MD

More information

A Scalable Adaptive Mesh Refinement Framework For Parallel Astrophysics Applications

A Scalable Adaptive Mesh Refinement Framework For Parallel Astrophysics Applications A Scalable Adaptive Mesh Refinement Framework For Parallel Astrophysics Applications James Bordner, Michael L. Norman San Diego Supercomputer Center University of California, San Diego 15th SIAM Conference

More information

Hybrid MPI + OpenMP Approach to Improve the Scalability of a Phase-Field-Crystal Code

Hybrid MPI + OpenMP Approach to Improve the Scalability of a Phase-Field-Crystal Code Hybrid MPI + OpenMP Approach to Improve the Scalability of a Phase-Field-Crystal Code Reuben D. Budiardja reubendb@utk.edu ECSS Symposium March 19 th, 2013 Project Background Project Team (University of

More information

Load Balancing and Data Migration in a Hybrid Computational Fluid Dynamics Application

Load Balancing and Data Migration in a Hybrid Computational Fluid Dynamics Application Load Balancing and Data Migration in a Hybrid Computational Fluid Dynamics Application Esteban Meneses Patrick Pisciuneri Center for Simulation and Modeling (SaM) University of Pittsburgh University of

More information

Using Hybrid Parallelism to Improve Memory Use in the Uintah Framework

Using Hybrid Parallelism to Improve Memory Use in the Uintah Framework Using Hybrid Parallelism to Improve Memory Use in the Uintah Framework Qingyu Meng Scientific Computing and Imaging Institute University of Utah Salt lake City, UT 84112 USA qymeng@cs.utah.edu Martin Berzins

More information

CUDA. Fluid simulation Lattice Boltzmann Models Cellular Automata

CUDA. Fluid simulation Lattice Boltzmann Models Cellular Automata CUDA Fluid simulation Lattice Boltzmann Models Cellular Automata Please excuse my layout of slides for the remaining part of the talk! Fluid Simulation Navier Stokes equations for incompressible fluids

More information

Algorithms, System and Data Centre Optimisation for Energy Efficient HPC

Algorithms, System and Data Centre Optimisation for Energy Efficient HPC 2015-09-14 Algorithms, System and Data Centre Optimisation for Energy Efficient HPC Vincent Heuveline URZ Computing Centre of Heidelberg University EMCL Engineering Mathematics and Computing Lab 1 Energy

More information

Scalable Parallel AMR for the Uintah Multiphysics Code

Scalable Parallel AMR for the Uintah Multiphysics Code Chapter 4 Scalable Parallel AMR for the Uintah Multiphysics Code Justin Luitjens SCI Institute, University of Utah Bryan Worthen SCI Institute, University of Utah Martin Berzins SCI Institute, University

More information

Center Extreme Scale CS Research

Center Extreme Scale CS Research Center Extreme Scale CS Research Center for Compressible Multiphase Turbulence University of Florida Sanjay Ranka Herman Lam Outline 10 6 10 7 10 8 10 9 cores Parallelization and UQ of Rocfun and CMT-Nek

More information

Adaptive Mesh Astrophysical Fluid Simulations on GPU. San Jose 10/2/2009 Peng Wang, NVIDIA

Adaptive Mesh Astrophysical Fluid Simulations on GPU. San Jose 10/2/2009 Peng Wang, NVIDIA Adaptive Mesh Astrophysical Fluid Simulations on GPU San Jose 10/2/2009 Peng Wang, NVIDIA Overview Astrophysical motivation & the Enzo code Finite volume method and adaptive mesh refinement (AMR) CUDA

More information

GAMER : a GPU-accelerated Adaptive-MEsh-Refinement Code for Astrophysics GPU 與自適性網格於天文模擬之應用與效能

GAMER : a GPU-accelerated Adaptive-MEsh-Refinement Code for Astrophysics GPU 與自適性網格於天文模擬之應用與效能 GAMER : a GPU-accelerated Adaptive-MEsh-Refinement Code for Astrophysics GPU 與自適性網格於天文模擬之應用與效能 Hsi-Yu Schive ( 薛熙于 ), Tzihong Chiueh ( 闕志鴻 ), Yu-Chih Tsai ( 蔡御之 ), Ui-Han Zhang ( 張瑋瀚 ) Graduate Institute

More information

Experiences with ENZO on the Intel R Many Integrated Core (Intel MIC) Architecture

Experiences with ENZO on the Intel R Many Integrated Core (Intel MIC) Architecture Experiences with ENZO on the Intel R Many Integrated Core (Intel MIC) Architecture 1 Introduction Robert Harkness National Institute for Computational Sciences Oak Ridge National Laboratory The National

More information

Requirements of Load Balancing Algorithm

Requirements of Load Balancing Algorithm LOAD BALANCING Programs and algorithms as graphs Geometric Partitioning Graph Partitioning Recursive Graph Bisection partitioning Recursive Spectral Bisection Multilevel Graph partitioning Hypergraph Partitioning

More information

ALE and AMR Mesh Refinement Techniques for Multi-material Hydrodynamics Problems

ALE and AMR Mesh Refinement Techniques for Multi-material Hydrodynamics Problems ALE and AMR Mesh Refinement Techniques for Multi-material Hydrodynamics Problems A. J. Barlow, AWE. ICFD Workshop on Mesh Refinement Techniques 7th December 2005 Acknowledgements Thanks to Chris Powell,

More information

CHRONO::HPC DISTRIBUTED MEMORY FLUID-SOLID INTERACTION SIMULATIONS. Felipe Gutierrez, Arman Pazouki, and Dan Negrut University of Wisconsin Madison

CHRONO::HPC DISTRIBUTED MEMORY FLUID-SOLID INTERACTION SIMULATIONS. Felipe Gutierrez, Arman Pazouki, and Dan Negrut University of Wisconsin Madison CHRONO::HPC DISTRIBUTED MEMORY FLUID-SOLID INTERACTION SIMULATIONS Felipe Gutierrez, Arman Pazouki, and Dan Negrut University of Wisconsin Madison Support: Rapid Innovation Fund, U.S. Army TARDEC ASME

More information

Improving the Performance of Uintah: A Large-Scale Adaptive Meshing Computational Framework

Improving the Performance of Uintah: A Large-Scale Adaptive Meshing Computational Framework Improving the Performance of Uintah: A Large-Scale Adaptive Meshing Computational Framework Justin Luitjens School of Computing University of Utah Salt Lake City, Utah 84112 luitjens@cs.utah.edu Martin

More information

Demonstrating GPU Code Portability and Scalability for Radiative Heat Transfer Computations

Demonstrating GPU Code Portability and Scalability for Radiative Heat Transfer Computations Demonstrating GPU Code Portability and Scalability for Radiative Heat Transfer Computations Brad Peterson, Alan Humphrey, John Holmen Todd Harman, Martin Berzins Scientific Computing and Imaging Institute

More information

Advanced Parallel Programming. Is there life beyond MPI?

Advanced Parallel Programming. Is there life beyond MPI? Advanced Parallel Programming Is there life beyond MPI? Outline MPI vs. High Level Languages Declarative Languages Map Reduce and Hadoop Shared Global Address Space Languages Charm++ ChaNGa ChaNGa on GPUs

More information

Lecture 13: Memory Consistency. + a Course-So-Far Review. Parallel Computer Architecture and Programming CMU , Spring 2013

Lecture 13: Memory Consistency. + a Course-So-Far Review. Parallel Computer Architecture and Programming CMU , Spring 2013 Lecture 13: Memory Consistency + a Course-So-Far Review Parallel Computer Architecture and Programming Today: what you should know Understand the motivation for relaxed consistency models Understand the

More information

Multigrid Solvers in CFD. David Emerson. Scientific Computing Department STFC Daresbury Laboratory Daresbury, Warrington, WA4 4AD, UK

Multigrid Solvers in CFD. David Emerson. Scientific Computing Department STFC Daresbury Laboratory Daresbury, Warrington, WA4 4AD, UK Multigrid Solvers in CFD David Emerson Scientific Computing Department STFC Daresbury Laboratory Daresbury, Warrington, WA4 4AD, UK david.emerson@stfc.ac.uk 1 Outline Multigrid: general comments Incompressible

More information

Multigrid Pattern. I. Problem. II. Driving Forces. III. Solution

Multigrid Pattern. I. Problem. II. Driving Forces. III. Solution Multigrid Pattern I. Problem Problem domain is decomposed into a set of geometric grids, where each element participates in a local computation followed by data exchanges with adjacent neighbors. The grids

More information

Strong Scaling for Molecular Dynamics Applications

Strong Scaling for Molecular Dynamics Applications Strong Scaling for Molecular Dynamics Applications Scaling and Molecular Dynamics Strong Scaling How does solution time vary with the number of processors for a fixed problem size Classical Molecular Dynamics

More information

How to Optimize Geometric Multigrid Methods on GPUs

How to Optimize Geometric Multigrid Methods on GPUs How to Optimize Geometric Multigrid Methods on GPUs Markus Stürmer, Harald Köstler, Ulrich Rüde System Simulation Group University Erlangen March 31st 2011 at Copper Schedule motivation imaging in gradient

More information

Introduction to C omputational F luid Dynamics. D. Murrin

Introduction to C omputational F luid Dynamics. D. Murrin Introduction to C omputational F luid Dynamics D. Murrin Computational fluid dynamics (CFD) is the science of predicting fluid flow, heat transfer, mass transfer, chemical reactions, and related phenomena

More information

Generating Efficient Data Movement Code for Heterogeneous Architectures with Distributed-Memory

Generating Efficient Data Movement Code for Heterogeneous Architectures with Distributed-Memory Generating Efficient Data Movement Code for Heterogeneous Architectures with Distributed-Memory Roshan Dathathri Thejas Ramashekar Chandan Reddy Uday Bondhugula Department of Computer Science and Automation

More information

Advances of parallel computing. Kirill Bogachev May 2016

Advances of parallel computing. Kirill Bogachev May 2016 Advances of parallel computing Kirill Bogachev May 2016 Demands in Simulations Field development relies more and more on static and dynamic modeling of the reservoirs that has come a long way from being

More information

European exascale applications workshop, Manchester, 11th and 12th October 2016 DLR TAU-Code - Application in INTERWinE

European exascale applications workshop, Manchester, 11th and 12th October 2016 DLR TAU-Code - Application in INTERWinE European exascale applications workshop, Manchester, 11th and 12th October 2016 DLR TAU-Code - Application in INTERWinE Thomas Gerhold, Barbara Brandfass, Jens Jägersküpper, DLR Christian Simmendinger,

More information

High performance computing and numerical modeling

High performance computing and numerical modeling High performance computing and numerical modeling Volker Springel Plan for my lectures Lecture 1: Collisional and collisionless N-body dynamics Lecture 2: Gravitational force calculation Lecture 3: Basic

More information

The ECMWF forecast model, quo vadis?

The ECMWF forecast model, quo vadis? The forecast model, quo vadis? by Nils Wedi European Centre for Medium-Range Weather Forecasts wedi@ecmwf.int contributors: Piotr Smolarkiewicz, Mats Hamrud, George Mozdzynski, Sylvie Malardel, Christian

More information

From Physics Model to Results: An Optimizing Framework for Cross-Architecture Code Generation

From Physics Model to Results: An Optimizing Framework for Cross-Architecture Code Generation From Physics Model to Results: An Optimizing Framework for Cross-Architecture Code Generation Erik Schnetter, Perimeter Institute with M. Blazewicz, I. Hinder, D. Koppelman, S. Brandt, M. Ciznicki, M.

More information

PhD University of Utah, Salt Lake City, UT Computer Science, expected December 2018

PhD University of Utah, Salt Lake City, UT Computer Science, expected December 2018 Alan Humphrey CONTACT INFORMATION ADDRESS: Scientific Computing and Imaging Institute 72 South Central Campus Drive, Rm 4819 University of Utah, Salt Lake City, UT 84112 PHONE: (801) 585-5090 (office)

More information

What is DARMA? DARMA is a C++ abstraction layer for asynchronous many-task (AMT) runtimes.

What is DARMA? DARMA is a C++ abstraction layer for asynchronous many-task (AMT) runtimes. DARMA Janine C. Bennett, Jonathan Lifflander, David S. Hollman, Jeremiah Wilke, Hemanth Kolla, Aram Markosyan, Nicole Slattengren, Robert L. Clay (PM) PSAAP-WEST February 22, 2017 Sandia National Laboratories

More information

ACCELERATING CFD AND RESERVOIR SIMULATIONS WITH ALGEBRAIC MULTI GRID Chris Gottbrath, Nov 2016

ACCELERATING CFD AND RESERVOIR SIMULATIONS WITH ALGEBRAIC MULTI GRID Chris Gottbrath, Nov 2016 ACCELERATING CFD AND RESERVOIR SIMULATIONS WITH ALGEBRAIC MULTI GRID Chris Gottbrath, Nov 2016 Challenges What is Algebraic Multi-Grid (AMG)? AGENDA Why use AMG? When to use AMG? NVIDIA AmgX Results 2

More information

Intro to Parallel Computing

Intro to Parallel Computing Outline Intro to Parallel Computing Remi Lehe Lawrence Berkeley National Laboratory Modern parallel architectures Parallelization between nodes: MPI Parallelization within one node: OpenMP Why use parallel

More information

HPC and IT Issues Session Agenda. Deployment of Simulation (Trends and Issues Impacting IT) Mapping HPC to Performance (Scaling, Technology Advances)

HPC and IT Issues Session Agenda. Deployment of Simulation (Trends and Issues Impacting IT) Mapping HPC to Performance (Scaling, Technology Advances) HPC and IT Issues Session Agenda Deployment of Simulation (Trends and Issues Impacting IT) Discussion Mapping HPC to Performance (Scaling, Technology Advances) Discussion Optimizing IT for Remote Access

More information

Lecture 6: Input Compaction and Further Studies

Lecture 6: Input Compaction and Further Studies PASI Summer School Advanced Algorithmic Techniques for GPUs Lecture 6: Input Compaction and Further Studies 1 Objective To learn the key techniques for compacting input data for reduced consumption of

More information

Load Balancing Techniques for Asynchronous Spacetime Discontinuous Galerkin Methods

Load Balancing Techniques for Asynchronous Spacetime Discontinuous Galerkin Methods Load Balancing Techniques for Asynchronous Spacetime Discontinuous Galerkin Methods Aaron K. Becker (abecker3@illinois.edu) Robert B. Haber Laxmikant V. Kalé University of Illinois, Urbana-Champaign Parallel

More information

Numerical Algorithms on Multi-GPU Architectures

Numerical Algorithms on Multi-GPU Architectures Numerical Algorithms on Multi-GPU Architectures Dr.-Ing. Harald Köstler 2 nd International Workshops on Advances in Computational Mechanics Yokohama, Japan 30.3.2010 2 3 Contents Motivation: Applications

More information

ParalleX. A Cure for Scaling Impaired Parallel Applications. Hartmut Kaiser

ParalleX. A Cure for Scaling Impaired Parallel Applications. Hartmut Kaiser ParalleX A Cure for Scaling Impaired Parallel Applications Hartmut Kaiser (hkaiser@cct.lsu.edu) 2 Tianhe-1A 2.566 Petaflops Rmax Heterogeneous Architecture: 14,336 Intel Xeon CPUs 7,168 Nvidia Tesla M2050

More information

Efficient Tridiagonal Solvers for ADI methods and Fluid Simulation

Efficient Tridiagonal Solvers for ADI methods and Fluid Simulation Efficient Tridiagonal Solvers for ADI methods and Fluid Simulation Nikolai Sakharnykh - NVIDIA San Jose Convention Center, San Jose, CA September 21, 2010 Introduction Tridiagonal solvers very popular

More information

PERFORMANCE PORTABILITY WITH OPENACC. Jeff Larkin, NVIDIA, November 2015

PERFORMANCE PORTABILITY WITH OPENACC. Jeff Larkin, NVIDIA, November 2015 PERFORMANCE PORTABILITY WITH OPENACC Jeff Larkin, NVIDIA, November 2015 TWO TYPES OF PORTABILITY FUNCTIONAL PORTABILITY PERFORMANCE PORTABILITY The ability for a single code to run anywhere. The ability

More information

Enzo-P / Cello. Formation of the First Galaxies. San Diego Supercomputer Center. Department of Physics and Astronomy

Enzo-P / Cello. Formation of the First Galaxies. San Diego Supercomputer Center. Department of Physics and Astronomy Enzo-P / Cello Formation of the First Galaxies James Bordner 1 Michael L. Norman 1 Brian O Shea 2 1 University of California, San Diego San Diego Supercomputer Center 2 Michigan State University Department

More information

Splotch: High Performance Visualization using MPI, OpenMP and CUDA

Splotch: High Performance Visualization using MPI, OpenMP and CUDA Splotch: High Performance Visualization using MPI, OpenMP and CUDA Klaus Dolag (Munich University Observatory) Martin Reinecke (MPA, Garching) Claudio Gheller (CSCS, Switzerland), Marzia Rivi (CINECA,

More information

TOOLS FOR IMPROVING CROSS-PLATFORM SOFTWARE DEVELOPMENT

TOOLS FOR IMPROVING CROSS-PLATFORM SOFTWARE DEVELOPMENT TOOLS FOR IMPROVING CROSS-PLATFORM SOFTWARE DEVELOPMENT Eric Kelmelis 28 March 2018 OVERVIEW BACKGROUND Evolution of processing hardware CROSS-PLATFORM KERNEL DEVELOPMENT Write once, target multiple hardware

More information

Fast Dynamic Load Balancing for Extreme Scale Systems

Fast Dynamic Load Balancing for Extreme Scale Systems Fast Dynamic Load Balancing for Extreme Scale Systems Cameron W. Smith, Gerrett Diamond, M.S. Shephard Computation Research Center (SCOREC) Rensselaer Polytechnic Institute Outline: n Some comments on

More information

Parallel High-Order Geometric Multigrid Methods on Adaptive Meshes for Highly Heterogeneous Nonlinear Stokes Flow Simulations of Earth s Mantle

Parallel High-Order Geometric Multigrid Methods on Adaptive Meshes for Highly Heterogeneous Nonlinear Stokes Flow Simulations of Earth s Mantle ICES Student Forum The University of Texas at Austin, USA November 4, 204 Parallel High-Order Geometric Multigrid Methods on Adaptive Meshes for Highly Heterogeneous Nonlinear Stokes Flow Simulations of

More information

1.2 Numerical Solutions of Flow Problems

1.2 Numerical Solutions of Flow Problems 1.2 Numerical Solutions of Flow Problems DIFFERENTIAL EQUATIONS OF MOTION FOR A SIMPLIFIED FLOW PROBLEM Continuity equation for incompressible flow: 0 Momentum (Navier-Stokes) equations for a Newtonian

More information

HPC Algorithms and Applications

HPC Algorithms and Applications HPC Algorithms and Applications Dwarf #5 Structured Grids Michael Bader Winter 2012/2013 Dwarf #5 Structured Grids, Winter 2012/2013 1 Dwarf #5 Structured Grids 1. dense linear algebra 2. sparse linear

More information

A4. Intro to Parallel Computing

A4. Intro to Parallel Computing Self-Consistent Simulations of Beam and Plasma Systems Steven M. Lund, Jean-Luc Vay, Rémi Lehe and Daniel Winklehner Colorado State U., Ft. Collins, CO, 13-17 June, 2016 A4. Intro to Parallel Computing

More information

ESPRESO ExaScale PaRallel FETI Solver. Hybrid FETI Solver Report

ESPRESO ExaScale PaRallel FETI Solver. Hybrid FETI Solver Report ESPRESO ExaScale PaRallel FETI Solver Hybrid FETI Solver Report Lubomir Riha, Tomas Brzobohaty IT4Innovations Outline HFETI theory from FETI to HFETI communication hiding and avoiding techniques our new

More information

Krishnan Suresh Associate Professor Mechanical Engineering

Krishnan Suresh Associate Professor Mechanical Engineering Large Scale FEA on the GPU Krishnan Suresh Associate Professor Mechanical Engineering High-Performance Trick Computations (i.e., 3.4*1.22): essentially free Memory access determines speed of code Pick

More information

ANSYS HPC. Technology Leadership. Barbara Hutchings ANSYS, Inc. September 20, 2011

ANSYS HPC. Technology Leadership. Barbara Hutchings ANSYS, Inc. September 20, 2011 ANSYS HPC Technology Leadership Barbara Hutchings barbara.hutchings@ansys.com 1 ANSYS, Inc. September 20, Why ANSYS Users Need HPC Insight you can t get any other way HPC enables high-fidelity Include

More information

Parallel Mesh Partitioning in Alya

Parallel Mesh Partitioning in Alya Available online at www.prace-ri.eu Partnership for Advanced Computing in Europe Parallel Mesh Partitioning in Alya A. Artigues a *** and G. Houzeaux a* a Barcelona Supercomputing Center ***antoni.artigues@bsc.es

More information

A Scalable Parallel LSQR Algorithm for Solving Large-Scale Linear System for Seismic Tomography

A Scalable Parallel LSQR Algorithm for Solving Large-Scale Linear System for Seismic Tomography 1 A Scalable Parallel LSQR Algorithm for Solving Large-Scale Linear System for Seismic Tomography He Huang, Liqiang Wang, Po Chen(University of Wyoming) John Dennis (NCAR) 2 LSQR in Seismic Tomography

More information

Scalable parallel regridding algorithms for block-structured adaptive mesh refinement

Scalable parallel regridding algorithms for block-structured adaptive mesh refinement CONCURRENCY AND COMPUTATION: PRACTICE AND EXPERIENCE Concurrency Computat.: Pract. Exper. 2000; 00:1 7 [Version: 2002/09/19 v2.02] Scalable parallel regridding algorithms for block-structured adaptive

More information

AmgX 2.0: Scaling toward CORAL Joe Eaton, November 19, 2015

AmgX 2.0: Scaling toward CORAL Joe Eaton, November 19, 2015 AmgX 2.0: Scaling toward CORAL Joe Eaton, November 19, 2015 Agenda Introduction to AmgX Current Capabilities Scaling V2.0 Roadmap for the future 2 AmgX Fast, scalable linear solvers, emphasis on iterative

More information

A Scalable GPU-Based Compressible Fluid Flow Solver for Unstructured Grids

A Scalable GPU-Based Compressible Fluid Flow Solver for Unstructured Grids A Scalable GPU-Based Compressible Fluid Flow Solver for Unstructured Grids Patrice Castonguay and Antony Jameson Aerospace Computing Lab, Stanford University GTC Asia, Beijing, China December 15 th, 2011

More information

Out-Of-Core Sort-First Parallel Rendering for Cluster-Based Tiled Displays

Out-Of-Core Sort-First Parallel Rendering for Cluster-Based Tiled Displays Out-Of-Core Sort-First Parallel Rendering for Cluster-Based Tiled Displays Wagner T. Corrêa James T. Klosowski Cláudio T. Silva Princeton/AT&T IBM OHSU/AT&T EG PGV, Germany September 10, 2002 Goals Render

More information

Ryan C. Hulguin TACC-Intel Highly Parallel Computing Symposium April 10th-11th, 2012 Austin, TX

Ryan C. Hulguin TACC-Intel Highly Parallel Computing Symposium April 10th-11th, 2012 Austin, TX Ryan C. Hulguin TACC-Intel Highly Parallel Computing Symposium April 10th-11th, 2012 Austin, TX Outline Introduction Knights Ferry Technical Specifications CFD Governing Equations Numerical Algorithm Solver

More information

Challenges Simulating Real Fuel Combustion Kinetics: The Role of GPUs

Challenges Simulating Real Fuel Combustion Kinetics: The Role of GPUs Challenges Simulating Real Fuel Combustion Kinetics: The Role of GPUs M. J. McNenly and R. A. Whitesides GPU Technology Conference March 27, 2014 San Jose, CA LLNL-PRES-652254! This work performed under

More information

Big Data Analytics Performance for Large Out-Of- Core Matrix Solvers on Advanced Hybrid Architectures

Big Data Analytics Performance for Large Out-Of- Core Matrix Solvers on Advanced Hybrid Architectures Procedia Computer Science Volume 51, 2015, Pages 2774 2778 ICCS 2015 International Conference On Computational Science Big Data Analytics Performance for Large Out-Of- Core Matrix Solvers on Advanced Hybrid

More information

Porting a parallel rotor wake simulation to GPGPU accelerators using OpenACC

Porting a parallel rotor wake simulation to GPGPU accelerators using OpenACC DLR.de Chart 1 Porting a parallel rotor wake simulation to GPGPU accelerators using OpenACC Melven Röhrig-Zöllner DLR, Simulations- und Softwaretechnik DLR.de Chart 2 Outline Hardware-Architecture (CPU+GPU)

More information

Dynamic Load Balancing for Weather Models via AMPI

Dynamic Load Balancing for Weather Models via AMPI Dynamic Load Balancing for Eduardo R. Rodrigues IBM Research Brazil edrodri@br.ibm.com Celso L. Mendes University of Illinois USA cmendes@ncsa.illinois.edu Laxmikant Kale University of Illinois USA kale@cs.illinois.edu

More information

Porting The Spectral Element Community Atmosphere Model (CAM-SE) To Hybrid GPU Platforms

Porting The Spectral Element Community Atmosphere Model (CAM-SE) To Hybrid GPU Platforms Porting The Spectral Element Community Atmosphere Model (CAM-SE) To Hybrid GPU Platforms http://www.scidacreview.org/0902/images/esg13.jpg Matthew Norman Jeffrey Larkin Richard Archibald Valentine Anantharaj

More information

arxiv: v1 [cs.ms] 8 Aug 2018

arxiv: v1 [cs.ms] 8 Aug 2018 ACCELERATING WAVE-PROPAGATION ALGORITHMS WITH ADAPTIVE MESH REFINEMENT USING THE GRAPHICS PROCESSING UNIT (GPU) XINSHENG QIN, RANDALL LEVEQUE, AND MICHAEL MOTLEY arxiv:1808.02638v1 [cs.ms] 8 Aug 2018 Abstract.

More information

HPC Application Porting to CUDA at BSC

HPC Application Porting to CUDA at BSC www.bsc.es HPC Application Porting to CUDA at BSC Pau Farré, Marc Jordà GTC 2016 - San Jose Agenda WARIS-Transport Atmospheric volcanic ash transport simulation Computer Applications department PELE Protein-drug

More information

Hybrid OpenMP-MPI Turbulent boundary Layer code over 32k cores

Hybrid OpenMP-MPI Turbulent boundary Layer code over 32k cores Hybrid OpenMP-MPI Turbulent boundary Layer code over 32k cores T/NT INTERFACE y/ x/ z/ 99 99 Juan A. Sillero, Guillem Borrell, Javier Jiménez (Universidad Politécnica de Madrid) and Robert D. Moser (U.

More information

NAMD at Extreme Scale. Presented by: Eric Bohm Team: Eric Bohm, Chao Mei, Osman Sarood, David Kunzman, Yanhua, Sun, Jim Phillips, John Stone, LV Kale

NAMD at Extreme Scale. Presented by: Eric Bohm Team: Eric Bohm, Chao Mei, Osman Sarood, David Kunzman, Yanhua, Sun, Jim Phillips, John Stone, LV Kale NAMD at Extreme Scale Presented by: Eric Bohm Team: Eric Bohm, Chao Mei, Osman Sarood, David Kunzman, Yanhua, Sun, Jim Phillips, John Stone, LV Kale Overview NAMD description Power7 Tuning Support for

More information