Скачать презентацию Data Fusion The Test and Evaluation Uses of Скачать презентацию Data Fusion The Test and Evaluation Uses of

268012dcedb2fd6b6d5f7090830e006e.ppt

  • Количество слайдов: 25

Data Fusion The Test and Evaluation Uses of Heterogeneous Computing 23 July 2010 Dan Data Fusion The Test and Evaluation Uses of Heterogeneous Computing 23 July 2010 Dan M. Davis ddavis@isi. edu Approved for public release; distribution is unlimited. (310)448 -8434

Co-Authors Prof. Robert F. Lucas and Gene Wagenbreth Information Sciences Institute University of Southern Co-Authors Prof. Robert F. Lucas and Gene Wagenbreth Information Sciences Institute University of Southern California Marina del Rey, California 90292 { rflucas, genew } @isi. edu

Overview • Configuration and Design Considerations • GPU Training and Algorithmic Programming • Current Overview • Configuration and Design Considerations • GPU Training and Algorithmic Programming • Current Contributions and Research Productivity • Robustness and Utility of GPGPU Cluster • Plans and Opportunities for Cluster • Lessons Learned regarding GPGPUs

Thesis Heterogeneous Computing (GPGPUs, FPGAs, STI Cells, …) holds promise for the future FMS Thesis Heterogeneous Computing (GPGPUs, FPGAs, STI Cells, …) holds promise for the future FMS and T&E have a need for HPC In 2007 HPCMP awarded a 256 -Node, GPGPUEnhanced Linux Cluster Joshua, to JFCOM This asset has proven stable and useful Many of the useful functions of GPGPUs will be applicable in the T&E community

Joshua GPGPU-Enhanced Linux Cluster at JFCOM J 9/J 7 Machine Room Suffolk Virginia Joshua GPGPU-Enhanced Linux Cluster at JFCOM J 9/J 7 Machine Room Suffolk Virginia

JFCOM as an GPGPU User ● U. S. Joint Forces Command, Norfolk, Virginia • JFCOM as an GPGPU User ● U. S. Joint Forces Command, Norfolk, Virginia • One of Do. D’s combatant commands • Key role in transforming defense capabilities • Current JFCOM Commander: Gen James Mattis, USMC ● Two JFCOM Directorates using agent-based simulations • J 7 - Training - trains forces develops doctrine leads training requirements analysis provides an interoperable training environment • J 9 - Concept Development and Experimentation - develops innovative joint concepts and capabilities - provides proven solutions to problems facing the joint force ● Simulations are typically Gen. Ser Secret and characterized by: • Interactive use by hundreds of personnel • Distributed trans-continentally, but must be real time • Vast majority of users at the terminals are uniformed warfighters

Simulation Federates ● Agent-based models use rules for entity behavior • Autonomous-agent entities • Simulation Federates ● Agent-based models use rules for entity behavior • Autonomous-agent entities • Can be Human-In-The-Loop (HITL) and run in real time • Large compute clusters required to run large-scale simulations ● Standard interface is HLA RTI communication (IEEE 1516) • Supplanted to old DIS • Publish/Subscribe model • USC/Caltech Software Routers scale better ● Common Codes in use at JFCOM: • • Joint Semi-Automated Forces (JSAF) “Culture”, stripped-down civilian instantiation of JSAF Simulating Location & Attack of Mobile Enemy Missiles (SLAMEM) One. SAF

GPGPU Justification NEED - 24 x 7 x 365 enhanced, distributed and scalable compute GPGPU Justification NEED - 24 x 7 x 365 enhanced, distributed and scalable compute resources to enable joint warfighters at JFCOM … to develop, explore, test, and validate 21 st century battlespace concepts … to enhance global-scale, computergenerated military experimentation by sustaining more than 2, 000 entities on appropriate terrain with valid phenomenology. APPROACH – Enable further growth in entity count, entity complexity, and environmental/Infrastructure settings by employing large Linux cluster with General Purpose GPUs (GPGPU) on each node to aid in line-of-sight, route planning, plume representation, all capable of running faster than real time. CHALLENGES – Effectively implementing Hardware configuration to provide stable and useful platform, motivate/train operators to utilize GPGPUs, and program simulations to take advantage of GPGPUs

Cluster Configuration as Delivered • 256 Nodes - (2) AMD Santa Rosa 2220 2. Cluster Configuration as Delivered • 256 Nodes - (2) AMD Santa Rosa 2220 2. 8 GHz dual-core processors GPUs - (1) NVIDIA 8800 Video Card Node Chassis - 4 U chassis Memory - 16 GB DIMM DDR 2 667 per node • Gig. E Inter-node Communications • Delivery to: Joint Advanced Tactics and Training Laboratory (JATTL) in Suffolk, VA

Perspective: Entity Growth vs. Time 10, 000 Number and Complexity of JSAF Entities SCALE Perspective: Entity Growth vs. Time 10, 000 Number and Complexity of JSAF Entities SCALE and FIDELITY Experiments continue to require orders of magnitude larger & more complex battlespaces 1, 400, 00 1, 000 DHPI GPUEnhanced Cluster DC Clusters at MHPCC & ASCMSRC SPP Proof of Principle DARPA / Caltech 250, 000 107, 000 12, 000 3, 600 UE 98 -1 (1997) SAF Express (1997) J 9901 (1999) 50, 000 AO-00 (2000) JSAF/SPP Urban Resolve (2004) JSAF/SPP Tests (2004) JSAF/SPP Capability Joshua (2006) (2008)

Why GPUs? ● GPU performance can be 100 X hosts • Don’t forget Prof. Why GPUs? ● GPU performance can be 100 X hosts • Don’t forget Prof. Gene Amdahl, 2 -3 X typical • This differential is expected to grow ● Early One. SAF work (UNC & SAIC) • • Line of Sight Route Finding Collision Detection Sparse Matrix Factorization (see RFLucas paper) ● ISI verified they’re also bottlenecks in JSAF ● New ideas for use in sensor scenario creation for new multi-spectral sensors

Route Planning Performance Impact Time Spent in Route Planning is Critical Bottleneck Route Planning Performance Impact Time Spent in Route Planning is Critical Bottleneck

Early GPU Programming ● Trained ISI staff with Sparse Matrix Solver ● Then examined Early GPU Programming ● Trained ISI staff with Sparse Matrix Solver ● Then examined JSAF kernels • Line-of-sight • Illumination • Route planning ● Route planning appeared easiest to integrate ● Route planning work published at I/ITSEC ● For this and other papers, see: http: //www. isi. edu/~ddavis/JESPP_Papers. html

CUDA Training PET Courses ● Dr. David Pratt conceived and organized • HPCMP FAPOC CUDA Training PET Courses ● Dr. David Pratt conceived and organized • HPCMP FAPOC for FMS ● Location & Dates: • SAIC facility Suffolk VA, 23 - 25 October 2007 • ISI Marina del Rey 21 - 23 October 2008 • UCSD San Diego 5 – 6 March 2009 ● Attendees: total ~ 60 HPCMP users ● Also taught at USC as part of Parallel Programming Class

Views of CUDA Classes in Suffolk Virginia Views of CUDA Classes in Suffolk Virginia

Typical Problem at JFCOM The Joint Force Commander (JFC) needs to integrate and focus Typical Problem at JFCOM The Joint Force Commander (JFC) needs to integrate and focus collection assets for persistent surveillance. Joint Integrated Persistent Surveillance (JIPS) Goals : Improving and integrating system Developing Tactics, Techniques and Procedures (TTP’s) Improving all of the Concept of Operations (CONOPS) Maximizing tipping, cueing and communications. Using sensors to achieve persistence Improving doctrine, organization, and TTPs Enabling JFC to better command support operations by: (1) effective capability apportionment and management, (2) timely and responsive analytic support (3) fast, reliable tactical Command, Control, Communications (C 3) Enhancing use, coordination and optimization of ISR assets

JIPS User Interface JIPS User Interface

Experimental Schedule Experimental Schedule

Benefits of GPGPU Computing Joshua has provided many benefits; some are not easily quantified Benefits of GPGPU Computing Joshua has provided many benefits; some are not easily quantified Training, analysis or evaluation in cities otherwise off-limits due to: security issues public resistance to combat troops in their city diplomatic about U. S. interest in cities of potential conflict Joshua does save personnel costs, e. g. Army Division costs ~ $20 M per day. DHPI cluster can runs such a program using only ~100 technicians Cost saving may be ~$19. 5 M each day. Good visibility with the leadership elite: Congressional visits Lieutenant General noted that it was probably the only time in his career he would have an opportunity to command so large a unit 1, 500 soldiers across the country participated, all connected by DREN to the cluster Joshua in Suffolk.

GPGPU Technical Merit ● ● All challenges in the proposal fully met Joshua remains GPGPU Technical Merit ● ● All challenges in the proposal fully met Joshua remains deployed and in service Two million entity goal exceeded (by factor of five!) Capability of GPU demonstrated • Developers trained to use GPUs • Route planning kernel implemented • Other research underway ● Joshua has changed the J 9 culture • New code being developed using client/server model • J 9 leadership now have ownership stake in HPC concepts

GPGPU Computational Merit ● JFCOM FMS requirements are uniquely military • Modeling of Do. GPGPU Computational Merit ● JFCOM FMS requirements are uniquely military • Modeling of Do. D operations in urban terrain • Users are most often uniformed warfighters • Recipients of research benefits are in the field today ● Needed for a large, heterogeneous ensemble of SAFs ● Cluster provides stability and mesh provides utility ● Nationally recognized research challenges • • Scalable interest management to bound messages Scaling individual behavior models Mining distributed data logs to analyze results More than 31 papers in competitive conferences/journals

GPGPU Current Progress ● ● ● Deployed and accepted at JFCOM In use on GPGPU Current Progress ● ● ● Deployed and accepted at JFCOM In use on all major J 9 experiments In use daily during development spirals for events Exceeded technical goal of hosting 2 M entities Classification issues led to partitioning Joshua is now fully engaged in day-to-day simulation experiments at JFCOM • Running ensembles of SLAMEM simulations ● Ops-tempo was expected to continue and increase • Human-in-the-loop experiments in FY 10

Summary Appropriateness ● Dedicated system was required • classified • interactive use • development Summary Appropriateness ● Dedicated system was required • classified • interactive use • development not amenable to batch processing ● Linux cluster • users have adapted easily and use constantly • design and use based on experience with DC clusters • current SAFs need only Low-cost Gig. E network ● Joshua has met JFCOMs requirements • in service creating data for JIPS • available for new directions

New Capabilities for T&E Paper in Real Time Hyper-Spectral Other T&E Uses Any line New Capabilities for T&E Paper in Real Time Hyper-Spectral Other T&E Uses Any line of sight calcualtions Equation-based CFD Signals Processing Matrix multiply

Research Funded by JFCOM and AFRL This material is based on research sponsored by Research Funded by JFCOM and AFRL This material is based on research sponsored by the U. S. Joint Forces Command via a contract with the Lockheed Martin Corporation and Sim. IS, Inc. , and on research sponsored by the Air Force Research Laboratory under agreement numbers F 30602 -02 -C-0213 and FA 8750 -05 -2 -0204. The U. S. Government is authorized to reproduce and distribute reprints for Governmental purposes notwithstanding any copyright notation thereon. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of the U. S. Government. Approved for public release; distribution is unlimited.