Developing a System for Real-Time Numerical Simulation during Physical Experiments in a Wave Propagation Laboratory

Size: px
Start display at page:

Download "Developing a System for Real-Time Numerical Simulation during Physical Experiments in a Wave Propagation Laboratory"

Transcription

1 Developing a System for Real-Time Numerical Simulation during Physical Experiments in a Wave Propagation Laboratory Dr. Thomas Blum Prof. Stewart Greenhalgh Prof. Johan Robertsson Dr. Dirk van Manen Marlies Vasmel ETH-Zurich D-EDRW Geophysics March 27, 2014 Darren Schmidt (presenter) Lothar Wenzel NI Scientific Research & Big Physics

2 The International Team NI Team Barry Hutt Darren Schmidt Balazs Toth Lothar Wenzel ETH Team Christoph Baerlocher Dr. Thomas Blum Prof. Stewart Greenhalgh Heinrich Horstmeyer Prof. Johan Robertsson Dr. Dirk van Manen Marlies Vasmel ` www. 2

3 Problem: Existing Laboratory Limitations Size of physical experiments limit the study of longer wavelengths or lower frequencies Rock Boundaries of physical experiments introduce undesired reflections reflection 3

4 Goals A new lab needs to: Address existing limitations Avoid introducing new barriers Allow for applications outside of geophysics 4

5 Solution Virtualize the Physical Experiment Immerse the physical experiment in a live real-time 3D simulation much like a person wearing a virtual reality suit is plugged into a visual simulation. Simulated Environment Rock 5

6 Solution Virtualize the Physical Experiment Immerse the physical experiment in a live real-time 3D simulation much like a person wearing a virtual reality suit is plugged into a visual simulation. 6

7 Solution Virtualize the Physical Experiment This new system allows Investigation of larger wavelengths Full control over wave propagation Example: Noise cancellation for undesired reflections. 7

8 Solution Virtualize the Physical Experiment This new system allows Investigation of larger wavelengths Full control over wave propagation Example: Noise cancellation for undesired reflections. hear 8

9 Solution Virtualize the Physical Experiment This new system allows Investigation of larger wavelengths Full control over wave propagation Example: Noise cancellation for undesired reflections. cancel 9

10 Solution Virtualize the Physical Experiment This new system allows Investigation of larger wavelengths Full control over wave propagation Example: Noise cancellation for undesired reflections. cancel 10

11 How? The physical components of the experiment are: A 1.5m cube rock sample 1.5m 1.5m Rock 1.5m 11

12 How? The physical components of the experiment are: A 1.5m cube rock sample A 2m cube of water containing the rock sample 2m Water 2m 2m 12

13 How? The physical components of the experiment are: A 1.5m cube rock sample A 2m cube of water containing the rock sample sensors on or close to rock surface o Sensors are equally distributed across six recording surfaces o Each sensor measures pressure sensors 13

14 How? The physical components of the experiment are: A 1.5m cube rock sample A 2m cube of water containing the rock sample sensors on or close to rock surface o Sensors are equally distributed across six recording surfaces o Each sensor measures pressure actuators on or close to water surface o Actuators are equally distributed across six emitting surfaces o Each actuator generates acoustic waves actuators 14

15 How? The physical components of the experiment are: A 1.5m cube rock sample A 2m cube of water containing the rock sample sensors on or close to rock surface o Sensors are equally distributed across six recording surfaces o Each sensor measures pressure actuators on or close to water surface o Actuators are equally distributed across six emitting surfaces o Each actuator generates acoustic waves 15

16 How? The simulation part of the experiment involves: Acquisition: channels Data represents pressure 16-bit resolution 16

17 How? The simulation part of the experiment involves: Acquisition: channels Computation: Large PDE solver Based on a pre-computed Green s functions (two 1K x 1K x 250 matrices ) Formulation requires pressure and velocity data 17

18 How? The simulation part of the experiment involves: Acquisition: channels Computation: Large PDE solver Control: channels Response represents generator values 16-bit resolution 18

19 How? The simulation part of the experiment involves: Acquisition: channels Computation: Large PDE solver Control: channels Real-time constraint: 50 ms cycle time Dictated by the 20 khz sampling rate Includes a complete acquire compute control cycle 19

20 Why. Water? So the system has the Time to act 20

21 Slower Wave Propagation Why. Water? So the system has the Time to act Distance from recording to emitting surface = 25 cm Propagation time in water ~ 170 ms 21

22 Slower Wave Propagation Why. Water? So the system has the Time to act Ability to react 22

23 Slower Wave Propagation Why. Water? So the system has the Time to act Ability to react Impedance differential impacts actuator performance Wave power in water is stronger than air 23

24 Slower Wave Propagation Why. Water? So the system has the Time to act Ability to react Impedance differential impacts actuator performance Wave power in water is stronger than air 24

25 Why. Water? So the system has the Time to act Ability to react Knowledge to respond Wave propagation in water is well-understood Defined by computable Green s functions 25

26 Why. Water? So the system has the Time to act Ability to react Knowledge to respond Wave propagation in water is well-understood Defined by computable Green s functions 26

27 Why the Rock? Predicting wave propagation in water is easy Behavior is known Models are based on linear operators Many computational optimizations are possible 27

28 Why the Rock? Predicting wave propagation in water is easy Predicting wave propagation in rock is hard Behavior is under exploration Models involve inherently non-linear functions Computational complexity is high 28

29 Why the Rock? Predicting wave propagation in water is easy Predicting wave propagation in rock is hard As a result, this real-time system: Leaves the nonlinear behavior to Mother Nature Captures the results at the recording surfaces Manipulates the physical experiment using known models 29

30 Why Pressure & Velocity? Dictated by the physics (i.e. the Green s functions) 30

31 Why Pressure & Velocity? Dictated by the physics (i.e. the Green s functions) velocity pressure 31

32 Why Pressure & Velocity? Dictated by the physics (i.e. the Green s functions) velocity pressure How Do We Get Velocity? 32

33 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Benefit o Compute velocity using six additional recording surfaces o New sensors measure pressure along the surface normal = sensor (for pressure) = sensor (for velocity) 33

34 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Benefit Work o Compute velocity using six additional recording surfaces o New sensors measure pressure along the surface normal o Compute the velocity using common difference method = sensor (for pressure) = sensor (for velocity) 34

35 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Benefit Work Issue o Compute velocity using six additional recording surfaces o New sensors measure pressure along the surface normal o Compute the velocity using common difference method o Sensor data is double and computation is increased. = sensor (for pressure) = sensor (for velocity) 35

36 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Option 2: Duplicate Surfaces Using Staggered Sensor Layout Benefit o Sensor data is not doubled o Gaps in one recording surface are filled in the other o Layout still leverages the surface normal = sensor (for pressure) = sensor (for velocity) 36

37 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Option 2: Duplicate Surfaces Using Staggered Sensor Layout Benefit Work o Sensor data is not doubled o Gaps in one recording surface are filled in the other o Layout still leverages the surface normal o Reconstruct all missing pressure data using interpolation = sensor (for pressure) = sensor (for velocity) = virtual sensor (for pressure) = virtual sensor (for velocity) 37

38 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Option 2: Duplicate Surfaces Using Staggered Sensor Layout Benefit Work Issue o Sensor data is not doubled o Gaps in one recording surface are filled in the other o Layout still leverages the surface normal o Reconstruct all missing pressure data using interpolation o Computation increases. = sensor (for pressure) = sensor (for velocity) = virtual sensor (for pressure) = virtual sensor (for velocity) 38

39 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Option 2: Duplicate Surfaces Using Staggered Sensor Layout Option 3: Optimize the Mathematical Model Benefit o Staggered layout is accounted for in the numerical simulation = sensor (for pressure) = sensor (for velocity) 39

40 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Option 2: Duplicate Surfaces Using Staggered Sensor Layout Option 3: Optimize the Mathematical Model Benefit Work o Staggered layout is accounted for in the numerical simulation o Velocity is implicitly computed via a new Green s function = sensor (for pressure) = sensor (for velocity) 40

41 Challenge 1 How Do We Get Velocity? Option 1: Duplicate Recording Surfaces Option 2: Duplicate Surfaces Using Staggered Sensor Layout Option 3: Optimize the Mathematical Model Benefit Work Issue o Staggered layout is accounted for in the numerical simulation o Velocity is implicitly computed via a new Green s function o No physical meaning to the new simulation model = sensor (for pressure) = sensor (for velocity) 41

42 Option 3 Optimize the Mathematical Model 42

43 Option 3 Optimize the Mathematical Model velocity pressure 43

44 Option 3 Optimize the Mathematical Model pressure 44

45 Option 3 Optimize the Mathematical Model pressure computed in advance 45

46 Option 3 Optimize the Mathematical Model The new Green s function: Avoids increasing the sensor data Reduces the computation by 50% 46

47 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth 47

48 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency o Is an issue for data movement within the system Ethernet USB PCI PCI Express Bandwidth (MB/s) 125 (Gigabit) 600 (SuperSpeed) (x16) Latency (ms) 1000 (Gigabit) <112 (SuperSpeed).7.7 (x4) NOTE: This table includes off-the-shelf bus & computer network technologies specific to instrumentation. 48

49 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency o Is an issue for data movement within the system o PCI Express bus offers high throughput and low latency o Transfers are synchronized using FPGA technology Ethernet USB PCI PCI Express Bandwidth (MB/s) 125 (Gigabit) 600 (SuperSpeed) (x16) Latency (ms) 1000 (Gigabit) <112 (SuperSpeed).7.7 (x4) NOTE: This table includes off-the-shelf bus & computer network technologies specific to instrumentation. 49

50 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth o Is an issue for the Green s functions computation 50

51 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth o Is an issue for the Green s functions computation o Pre-computed Green s functions data is large 1000 sensors & 1000 actuators 1K x 1K Experiment runs for 250 time steps

52 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth o Is an issue for the Green s functions computation o Pre-computed Green s functions data is large 1000 sensors & 1000 actuators 1K x 1K Experiment runs for 250 time steps {1K x (1K x 250)} sgemv ops 52

53 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth o Is an issue for the Green s functions computation o Pre-computed Green s functions data is large o Consume 4GB of memory Storage = 4GB 53

54 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth o Is an issue for the Green s functions computation o Pre-computed Green s functions data is large o Consume 4GB of memory o Computation requires 100TB/s memory bandwidth Estimated with data movement overhead of 10 ms System bandwidth = 100TB/s 54

55 Challenge 2 How to Simulate in Real-Time? Given a cycle time of 50 ms Latency Bandwidth o Is an issue for the Green s functions computation o Pre-computed Green s functions data is large o Consume 4GB of memory o Computation requires 100TB/s memory bandwidth o To perform the computation in real-time Leverage multiple NVIDIA Tesla K40 GPUs Connected via PCI Express Optimize the Green s functions implementation to reduce the number of GPUs required by an order of magnitude Telsa K40: 288GB/s 55

56 Real-Time 3D Simulator Components Emitting surface GPUs PXImc PCIe Switch MXI Gen2 MXI Gen2 Recording surface PXI FPGA LabVIEW Timing/Synch Network over PCIe GPU acceleration PCIe enclosures PCIe switches 56

57 Demo: System Simulation & Green s Functions Computation Simulation of the System Green s Functions Computation Sensor Data 57

58 Special Thanks NVIDIA Jerry Chen Cliff Woolley Duncan Poole One Stop Systems Jaan Mannik Dell Aron Bowan 58

59 References Vasmel et al (2013). Immersive experimentation in a wave propagation laboratory, J. Acoust. Soc. Am. 134, EL492-EL Big Physics at National Instrument s 59

What is PXImc? By Chetan Kapoor, PXI Product Manager National Instruments

What is PXImc? By Chetan Kapoor, PXI Product Manager National Instruments What is PXImc? By Chetan Kapoor, PXI Product Manager National Instruments Overview Modern day test and control systems are growing larger, more complex and more intricate. Most of these intricacies are

More information

Create Without Limits: Add the Power of User-Programmable FPGAs to Your Test Applications

Create Without Limits: Add the Power of User-Programmable FPGAs to Your Test Applications 1 Create Without Limits: Add the Power of User-Programmable FPGAs to Your Test Applications Farris Alhorr Business Development Manager RF & Wireless Comm farris.alhorr@ The Parameters of Instrumentation

More information

Scientific Instrumentation using NI Technology

Scientific Instrumentation using NI Technology Scientific Instrumentation using NI Technology Presented by, Raja Pillai Technical Consultant and Field Engineer National Instruments Australia Pty. Ltd. NI s Platform-Based Approach 2 The LabVIEW RIO

More information

PC-based data acquisition I

PC-based data acquisition I FYS3240 PC-based instrumentation and microcontrollers PC-based data acquisition I Spring 2016 Lecture #8 Bekkeng, 20.01.2016 General-purpose computer With a Personal Computer (PC) we mean a general-purpose

More information

How to Choose the Right Bus for Your Measurement System

How to Choose the Right Bus for Your Measurement System 1 How to Choose the Right Bus for Your Measurement System Overview When you have hundreds of different data acquisition (DAQ) devices to choose from on a wide variety of buses, it can be difficult to select

More information

What s New in Data Acquisition

What s New in Data Acquisition What s New in Data Acquisition NEW! New DAQ Products and Demos NEW! USB DAQ Multifunction Intelligent DAQ DAQ Software New DAQ Products and Demos NEW! USB DAQ Multifunction Intelligent DAQ DAQ Software

More information

A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator Servers

A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator Servers A PCIe Congestion-Aware Performance Model for Densely Populated Accelerator Servers Maxime Martinasso, Grzegorz Kwasniewski, Sadaf R. Alam, Thomas C. Schulthess, Torsten Hoefler Swiss National Supercomputing

More information

High Performance Embedded Applications. Raja Pillai Applications Engineering Specialist

High Performance Embedded Applications. Raja Pillai Applications Engineering Specialist High Performance Embedded Applications Raja Pillai Applications Engineering Specialist Agenda What is High Performance Embedded? NI s History in HPE FlexRIO Overview System architecture Adapter modules

More information

PXI Remote Control and System Expansion

PXI Remote Control and System Expansion Have a question? Contact Us. PRODUCT FLYER PXI Remote Control and System Expansion CONTENTS PXI Remote Control and System Expansion Components of a Remotely Controlled PXI System Choosing a Remote Control

More information

High-Value PXI Embedded Controller for Windows. High-Value Embedded Controllers for PXI Express NI PXI-8101, NI PXI NI PXIe-8101, NI PXIe-8102

High-Value PXI Embedded Controller for Windows. High-Value Embedded Controllers for PXI Express NI PXI-8101, NI PXI NI PXIe-8101, NI PXIe-8102 High-Value PXI Embedded Controller for Windows NI PXI-8101, NI PXI-8102 2.0 GHz single-core for PXI-8101, 1.9 GHz dual-core for PXI-8102 1 GB (1 x 1 GB DIMM) 800 MHz DDR2 RAM standard, 4 GB (1 x 4 GB DIMMs)

More information

Simplify System Complexity

Simplify System Complexity Simplify System Complexity With the new high-performance CompactRIO controller Fanie Coetzer Field Sales Engineer Northern South Africa 2 3 New control system CompactPCI MMI/Sequencing/Logging FieldPoint

More information

The Benefits of FPGA-Enabled Instruments in RF and Communications Test. Johan Olsson National Instruments Sweden AB

The Benefits of FPGA-Enabled Instruments in RF and Communications Test. Johan Olsson National Instruments Sweden AB The Benefits of FPGA-Enabled Instruments in RF and Communications Test Johan Olsson National Instruments Sweden AB 1 Agenda Introduction to FPGAs in test New FPGA-enabled test applications FPGA for test

More information

Mapping MPI+X Applications to Multi-GPU Architectures

Mapping MPI+X Applications to Multi-GPU Architectures Mapping MPI+X Applications to Multi-GPU Architectures A Performance-Portable Approach Edgar A. León Computer Scientist San Jose, CA March 28, 2018 GPU Technology Conference This work was performed under

More information

Introduction to LabVIEW and NI Hardware Platform

Introduction to LabVIEW and NI Hardware Platform Introduction to LabVIEW and NI Hardware Platform Corrie Botha Platform-Based Approach 2 With LabVIEW, You Can Program the Way You Think 3 With LabVIEW, You Can Program the Way You Think The graphical,

More information

How to validate your FPGA design using realworld

How to validate your FPGA design using realworld How to validate your FPGA design using realworld stimuli Daniel Clapham National Instruments ni.com Agenda Typical FPGA Design NIs approach to FPGA Brief intro into platform based approach RIO architecture

More information

Simplify System Complexity

Simplify System Complexity 1 2 Simplify System Complexity With the new high-performance CompactRIO controller Arun Veeramani Senior Program Manager National Instruments NI CompactRIO The Worlds Only Software Designed Controller

More information

Building a Next Generation Data Logging System

Building a Next Generation Data Logging System 1 Building a Next Generation Data Logging System Fanie Coetzer, Field Sales Engineer Northern South Africa Outline Introduction to the Next Generation of Data Logging Signals and Signal Conditioning Data

More information

Introduction to PXI. Jacques Cilliers. Applications Engineer National Instruments ni.com

Introduction to PXI. Jacques Cilliers. Applications Engineer National Instruments ni.com 1 Introduction to PXI Jacques Cilliers Applications Engineer National Instruments jacques.cilliers@ Traditional Instrumentation 4 The Expectations More Functionality Reduced Cost Smaller Footprint Accelerated

More information

NI Vision Platform. Radim ŠTEFAN. ni.com

NI Vision Platform. Radim ŠTEFAN.  ni.com NI Vision Platform Radim ŠTEFAN www./vision National Instruments Our Stability Revenue: $1.15B in 2012 Innovation: 18% re-invested to R&D Global Operations: Approximately 7,100 employees; operations in

More information

FYS Data acquisition & control. Introduction. Spring 2018 Lecture #1. Reading: RWI (Real World Instrumentation) Chapter 1.

FYS Data acquisition & control. Introduction. Spring 2018 Lecture #1. Reading: RWI (Real World Instrumentation) Chapter 1. FYS3240-4240 Data acquisition & control Introduction Spring 2018 Lecture #1 Reading: RWI (Real World Instrumentation) Chapter 1. Bekkeng 14.01.2018 Topics Instrumentation: Data acquisition and control

More information

DSP using Labview FPGA. T.J.Moir AUT University School of Engineering Auckland New-Zealand

DSP using Labview FPGA. T.J.Moir AUT University School of Engineering Auckland New-Zealand DSP using Labview FPGA T.J.Moir AUT University School of Engineering Auckland New-Zealand Limitations of a basic processor Despite all of the advancements we ve made in the world of processors, they still

More information

Modern State Space Representation Using ADwin DAQ Systems

Modern State Space Representation Using ADwin DAQ Systems Modern State Space Representation Using ADwin DAQ Systems ADwin Provides High Throughput and Simulink Integration CAS DataLoggers provided the ADwin data acquisition and control solution to an internationally-known

More information

I/O Devices. I/O Management Intro. Sample Data Rates. I/O Device Handling. Categories of I/O Devices (by usage)

I/O Devices. I/O Management Intro. Sample Data Rates. I/O Device Handling. Categories of I/O Devices (by usage) I/O Devices I/O Management Intro Chapter 5 There exists a large variety of I/O devices: Many of them with different properties They seem to require different interfaces to manipulate and manage them We

More information

Computer buses and interfaces

Computer buses and interfaces FYS3240-4240 Data acquisition & control Computer buses and interfaces Spring 2018 Lecture #7 Reading: RWI Ch7 and page 559 Bekkeng 14.02.2018 Abbreviations B = byte b = bit M = mega G = giga = 10 9 k =

More information

WP 14 and Timing Sync

WP 14 and Timing Sync WP 14 and Timing Sync Eiscat Technical meeting 20131105 Leif Johansson National Instruments Eiscat Syncronisation Signal vs. Time-Based Synchronization Signal-Based Share Physical Clocks / Triggers Time-Based

More information

PXI - An ideal platform for a variety of industrial applications

PXI - An ideal platform for a variety of industrial applications PXI - An ideal platform for a variety of industrial applications Overview This white paper explains in detail the PXI architecture and discusses the benefits PXI systems present in a variety of industrial

More information

Integrating Data Acquisition and Instrument Control with Your Scilab Scripts

Integrating Data Acquisition and Instrument Control with Your Scilab Scripts Integrating Data Acquisition and Instrument Control with Your Scilab Scripts Darcy Dement Marketing Director National Instruments France darcy.dement@ni.com Who We Are 800 Leaders in Computer-based Measurement

More information

Interconnect Your Future

Interconnect Your Future Interconnect Your Future Gilad Shainer 2nd Annual MVAPICH User Group (MUG) Meeting, August 2014 Complete High-Performance Scalable Interconnect Infrastructure Comprehensive End-to-End Software Accelerators

More information

The LabVIEW RIO Architecture and the Newest Member to the CompactRIO Family

The LabVIEW RIO Architecture and the Newest Member to the CompactRIO Family The LabVIEW RIO Architecture and the Newest Member to the CompactRIO Family Brett Burger Sr. Product Manager What is CompactRIO? IP Camera GigE Camera Analog Frame Grabber Analog Camera Motor Drivers and

More information

DAQ & Control with PXI. Murali Ravindran Senior Product Manager

DAQ & Control with PXI. Murali Ravindran Senior Product Manager DAQ & Control with PXI Murali Ravindran Senior Product Manager Agenda What is PXI? Trigger with PXI Multicore Programming DAQ & Control with FPGA Instrumentation Timeline 1965 1987 1995 1997 Photo Courtesy

More information

I/O Management Intro. Chapter 5

I/O Management Intro. Chapter 5 I/O Management Intro Chapter 5 1 Learning Outcomes A high-level understanding of the properties of a variety of I/O devices. An understanding of methods of interacting with I/O devices. An appreciation

More information

LabView Add-On Manual

LabView Add-On Manual Specialty Plants Specialty Experiment: PIV-plus-Feedforward Position Control Magnetic Levitation Plant LabView Add-On Manual Table of Contents 1. Objectives...1 2. Prerequisites...1 3. References...1 4.

More information

NI Technical Symposium 2016

NI Technical Symposium 2016 NI Technical Symposium 2016 How can Modular Instruments reduce the Cost of Test? Fanie Coetzer Field Sales Engineer National Instruments Box Instruments 3 Box Inserting New Technology 1 2 3 4 5 6 7 8 5

More information

Advanced NI-DAQmx Programming Techniques with LabVIEW

Advanced NI-DAQmx Programming Techniques with LabVIEW Advanced NI-DAQmx Programming Techniques with LabVIEW Agenda Understanding Your Hardware Data Acquisition Systems Data Acquisition Device Subsystems Advanced Programming with NI-DAQmx Understanding Your

More information

GeoImaging Accelerator Pansharpen Test Results. Executive Summary

GeoImaging Accelerator Pansharpen Test Results. Executive Summary Executive Summary After demonstrating the exceptional performance improvement in the orthorectification module (approximately fourteen-fold see GXL Ortho Performance Whitepaper), the same approach has

More information

Power Systems AC922 Overview. Chris Mann IBM Distinguished Engineer Chief System Architect, Power HPC Systems December 11, 2017

Power Systems AC922 Overview. Chris Mann IBM Distinguished Engineer Chief System Architect, Power HPC Systems December 11, 2017 Power Systems AC922 Overview Chris Mann IBM Distinguished Engineer Chief System Architect, Power HPC Systems December 11, 2017 IBM POWER HPC Platform Strategy High-performance computer and high-performance

More information

Gen-Z Memory-Driven Computing

Gen-Z Memory-Driven Computing Gen-Z Memory-Driven Computing Our vision for the future of computing Patrick Demichel Distinguished Technologist Explosive growth of data More Data Need answers FAST! Value of Analyzed Data 2005 0.1ZB

More information

Marine Acoustic Acquisition System

Marine Acoustic Acquisition System Omiga Technology Ltd was founded in 2000 providing bespoke software and hardware solutions for high speed data acquisition systems and data analysis. The majority of solutions provided are based on National

More information

New Software-Designed Instruments

New Software-Designed Instruments 1 New Software-Designed Instruments Nicholas Haripersad Field Applications Engineer National Instruments South Africa Agenda What Is a Software-Designed Instrument? Why Software-Designed Instrumentation?

More information

Simulating Shallow Water on GPUs Programming of Heterogeneous Systems in Physics

Simulating Shallow Water on GPUs Programming of Heterogeneous Systems in Physics Simulating Shallow Water on GPUs Programming of Heterogeneous Systems in Physics Martin Pfeiffer (m.pfeiffer@uni-jena.de) Friedrich Schiller University Jena 06.10.2011 Simulating Shallow Water on GPUs

More information

Software-Defined Test Fundamentals. Understanding the Architecture of Modular, High-Performance Test Systems

Software-Defined Test Fundamentals. Understanding the Architecture of Modular, High-Performance Test Systems Software-Defined Test Fundamentals Understanding the Architecture of Modular, High-Performance Test Systems Contents Executive Summary 4 Architecture Layer No. 5: System Management/Test Executive 5 Architecture

More information

New Technologies in Data Acquisition

New Technologies in Data Acquisition New Technologies in Data Acquisition Agenda Introduction to PC Based Data Acquisition New Technologies in Data Acquisition 1. New Bus Technologies 2. Signal Conditioning, Timing and Synchronization Introduction

More information

An Experimental Study of Network Performance Impact of Increased Latency in SDR

An Experimental Study of Network Performance Impact of Increased Latency in SDR An Experimental Study of Network Performance Impact of Increased Latency in SDR Thomas Schmid Oussama Sekkat Mani B. Srivastava - Wintech workshop was started with the Keynote from Eric Blossom on GNU

More information

Automating Calibration Systems

Automating Calibration Systems Automating Calibration Systems Dave Skinner Global Services, Business Manager Agenda Why Automate? Automation Solutions Challenges with Automating Calibration Calibration Automation Architecture Calibration

More information

Ettus Research Update

Ettus Research Update Ettus Research Update Matt Ettus Ettus Research GRCon13 Outline 1 Introduction 2 Recent New Products 3 Third Generation Introduction Who am I? Core GNU Radio contributor since 2001 Designed

More information

Building NVLink for Developers

Building NVLink for Developers Building NVLink for Developers Unleashing programmatic, architectural and performance capabilities for accelerated computing Why NVLink TM? Simpler, Better and Faster Simplified Programming No specialized

More information

LabWindows /CVI Using the Latest ANSI C Tools for High-Performance Automated Test

LabWindows /CVI Using the Latest ANSI C Tools for High-Performance Automated Test LabWindows /CVI Using the Latest ANSI C Tools for High-Performance Automated Test Agenda Introduction to LabWindows/CVI and modular instruments Developing modular instrument applications Hybrid systems

More information

VIBRATION ISOLATION USING A MULTI-AXIS ROBOTIC PLATFORM G.

VIBRATION ISOLATION USING A MULTI-AXIS ROBOTIC PLATFORM G. VIBRATION ISOLATION USING A MULTI-AXIS ROBOTIC PLATFORM G. Satheesh Kumar, Y. G. Srinivasa and T. Nagarajan Precision Engineering and Instrumentation Laboratory Department of Mechanical Engineering Indian

More information

International Journal of Advance Engineering and Research Development. Flow Control Loop Analysis for System Modeling & Identification

International Journal of Advance Engineering and Research Development. Flow Control Loop Analysis for System Modeling & Identification Scientific Journal of Impact Factor(SJIF): 3.134 e-issn(o): 2348-4470 p-issn(p): 2348-6406 International Journal of Advance Engineering and Research Development Volume 2,Issue 5, May -2015 Flow Control

More information

genzconsortium.org Gen-Z Technology: Enabling Memory Centric Architecture

genzconsortium.org Gen-Z Technology: Enabling Memory Centric Architecture Gen-Z Technology: Enabling Memory Centric Architecture Why Gen-Z? Gen-Z Consortium 2017 2 Why Gen-Z? Gen-Z Consortium 2017 3 Why Gen-Z? Businesses Need to Monetize Data Big Data AI Machine Learning Deep

More information

S2C K7 Prodigy Logic Module Series

S2C K7 Prodigy Logic Module Series S2C K7 Prodigy Logic Module Series Low-Cost Fifth Generation Rapid FPGA-based Prototyping Hardware The S2C K7 Prodigy Logic Module is equipped with one Xilinx Kintex-7 XC7K410T or XC7K325T FPGA device

More information

IDD Aerospace. Cursor Control Device Human Factors and Technology Considerations for Flightdeck Applications

IDD Aerospace. Cursor Control Device Human Factors and Technology Considerations for Flightdeck Applications IDD Aerospace Cursor Control Device Human Factors and Technology Considerations for Flightdeck Applications Doug Daman Copyright 2009 Zodiac Aerospace All Rights Reserved 1 Introduction Competition is

More information

I/O Management Intro. Chapter 5

I/O Management Intro. Chapter 5 I/O Management Intro Chapter 5 1 Learning Outcomes A high-level understanding of the properties of a variety of I/O devices. An understanding of methods of interacting with I/O devices. 2 I/O Devices There

More information

Designing Next Generation Test Systems An In-Depth Developers Guide

Designing Next Generation Test Systems An In-Depth Developers Guide An In-Depth Developers Guide Designing Next Generation Test Systems An In-depth Developers Guide Contents Section 1 Executive Summary Chapter 1 Increasing Design Complexity...1-1 Shorter Product Development

More information

HCS1 Three-Axis Helmholtz Coil System

HCS1 Three-Axis Helmholtz Coil System HCS1 Three-Axis Helmholtz Coil System Innovation in Magnetic Measuring Instruments HCS1 Three-Axis Helmholtz Coil System HCS1 Three-Axis Helmholtz Coil System The system consists of three pairs of coils

More information

CMX18 FEATURES A 18-SLOT 3U PXI EXPRESS CHASSIS

CMX18 FEATURES A 18-SLOT 3U PXI EXPRESS CHASSIS 83-0062-000 15A D A T A S H E E T CMX18 18-SLOT 3U PXI EXPRESS CHASSIS FEATURES 18-slot PXI Express chassis with 1 system controller slot, 6 peripheral slots, 10 hybrid slots, and 1 timing slot High bandwidth

More information

Accelerating computation with FPGAs

Accelerating computation with FPGAs Accelerating computation with FPGAs Michael J. Flynn Maxeler Technologies and Stanford University M. J. Flynn Maxeler Technologies 1 Based on work done by my colleagues at Maxeler, especially Oskar Mencer,

More information

Porting CPU-based Multiprocessing Algorithms to GPU for Distributed Acoustic Sensing

Porting CPU-based Multiprocessing Algorithms to GPU for Distributed Acoustic Sensing GTC2014 S4470 Porting CPU-based Multiprocessing Algorithms to GPU for Distributed Acoustic Sensing Steve Jankly Halliburton Energy Services, Inc. Introduction Halliburton Halliburton is one of the world

More information

PMX09 DATA SHEET FEATURES A PORTABLE 9-SLOT 3U PXI EXPRESS CHASSIS WITH INTEGRATED DISPLAY AND KEYBOARD

PMX09 DATA SHEET FEATURES A PORTABLE 9-SLOT 3U PXI EXPRESS CHASSIS WITH INTEGRATED DISPLAY AND KEYBOARD 83-0039-000 14A DATA SHEET PMX09 PORTABLE 9-SLOT 3U PXI EXPRESS CHASSIS WITH INTEGRATED DISPLAY AND KEYBOARD FEATURES 9-slot PXI Express chassis with 1 system controller slot, 6 peripheral slots, 1 hybrid

More information

Adaptive-Mesh-Refinement Hydrodynamic GPU Computation in Astrophysics

Adaptive-Mesh-Refinement Hydrodynamic GPU Computation in Astrophysics Adaptive-Mesh-Refinement Hydrodynamic GPU Computation in Astrophysics H. Y. Schive ( 薛熙于 ) Graduate Institute of Physics, National Taiwan University Leung Center for Cosmology and Particle Astrophysics

More information

64-Channel, 16-Bit Simultaneous Sampling PMC Analog Input Board

64-Channel, 16-Bit Simultaneous Sampling PMC Analog Input Board 66-16AI64SSA/C 64-Channel, 16-Bit Simultaneous Sampling PMC Analog Board With 200 KSPS Sample Rate per Channel and 66 MHz PCI Support Available in PMC, PCI, cpci and PC104-Plus and PCI Express form factors

More information

HALF YEARLY EXAMINATIONS 2015/2016. Answer ALL questions showing your working Where necessary give your answers correct to 2 decimal places.

HALF YEARLY EXAMINATIONS 2015/2016. Answer ALL questions showing your working Where necessary give your answers correct to 2 decimal places. Track 3 GIRLS SECONDARY, MRIEHEL HALF YEARLY EXAMINATIONS 2015/2016 FORM: 4 PHYSICS Time: 1½ hrs Name: Class: Answer ALL questions showing your working Where necessary give your answers correct to 2 decimal

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

vs. GPU Performance Without the Answer University of Virginia Computer Engineering g Labs

vs. GPU Performance Without the Answer University of Virginia Computer Engineering g Labs Where is the Data? Why you Cannot Debate CPU vs. GPU Performance Without the Answer Chris Gregg and Kim Hazelwood University of Virginia Computer Engineering g Labs 1 GPUs and Data Transfer GPU computing

More information

Control Technology. motion controller and power amplifier

Control Technology. motion controller and power amplifier Control Technology motion controller and power amplifier Erik van Hilten Rik Prins National Instruments Agenda Controller, the central element Tools for controller design in drive systems: - in PC-based

More information

Tu Crossline Reconstruction Using Aliased 3D Deghosted Up- and Downgoing Wavefields

Tu Crossline Reconstruction Using Aliased 3D Deghosted Up- and Downgoing Wavefields Tu-04-10 Crossline Reconstruction Using Aliased 3D Deghosted Up- and Downgoing Wavefields D.-J. van Manen* (Schlumberger), M. Vassallo (WesternGeco), A.K. Özdemir (WesternGeco), A. Özbek (Schlumberger)

More information

DAQ SYSTEM FOR GYROLASER

DAQ SYSTEM FOR GYROLASER Bachir Bouhadef 1 Analog input channels : Icw, Iccw and Sagnac @ 5 khz. Control parameters ( >10 ) @ 1 Hz. Sagnac Icw Iccw 2 NI PXI-8106 RT : 2.16 GHz Intel Core 2 Duo T7400 With LabVIEW Real-Time 3 What

More information

Wavemaker Improvements at the University of New Orleans Towing Tank

Wavemaker Improvements at the University of New Orleans Towing Tank University of New Orleans ScholarWorks@UNO Ocean Waves Workshop Ocean Waves Workshop 2017 Dec 7th, 8:30 AM - 9:00 AM Wavemaker Improvements at the University of New Orleans Towing Tank Ryan D. Thiel University

More information

IBM Power AC922 Server

IBM Power AC922 Server IBM Power AC922 Server The Best Server for Enterprise AI Highlights More accuracy - GPUs access system RAM for larger models Faster insights - significant deep learning speedups Rapid deployment - integrated

More information

GPU Ultrasound Simulation and Volume Reconstruction

GPU Ultrasound Simulation and Volume Reconstruction GPU Ultrasound Simulation and Volume Reconstruction Athanasios Karamalis 1,2 Supervisor: Nassir Navab1 Advisor: Oliver Kutter1, Wolfgang Wein2 1Computer Aided Medical Procedures (CAMP), Technische Universität

More information

BlueDBM: An Appliance for Big Data Analytics*

BlueDBM: An Appliance for Big Data Analytics* BlueDBM: An Appliance for Big Data Analytics* Arvind *[ISCA, 2015] Sang-Woo Jun, Ming Liu, Sungjin Lee, Shuotao Xu, Arvind (MIT) and Jamey Hicks, John Ankcorn, Myron King(Quanta) BigData@CSAIL Annual Meeting

More information

Motor Control: Model-Based Design from Concept to Implementation on heterogeneous SoC FPGAs Alexander Schreiber, MathWorks

Motor Control: Model-Based Design from Concept to Implementation on heterogeneous SoC FPGAs Alexander Schreiber, MathWorks Motor Control: Model-Based Design from Concept to Implementation on heterogeneous SoC FPGAs Alexander Schreiber, MathWorks 2014 The MathWorks, Inc. 1 Some components of a production application Production

More information

Efficient CPU GPU data transfers CUDA 6.0 Unified Virtual Memory

Efficient CPU GPU data transfers CUDA 6.0 Unified Virtual Memory Institute of Computational Science Efficient CPU GPU data transfers CUDA 6.0 Unified Virtual Memory Juraj Kardoš (University of Lugano) July 9, 2014 Juraj Kardoš Efficient GPU data transfers July 9, 2014

More information

- Part 3 - Energy Aware Numerics. Vincent Heuveline

- Part 3 - Energy Aware Numerics. Vincent Heuveline - Part 3 - Energy Aware Numerics Vincent Heuveline 1 The Challenge 2 Exa2green Engineering Mathematics and Computing Lab UHEI Steinbeis Europa Zentrum SEZ Scientific Computing Group UHAM High Performance

More information

Architecture of an efficient MEMS final test system Dr. Martin Brucke, SPEKTRA Dresden

Architecture of an efficient MEMS final test system Dr. Martin Brucke, SPEKTRA Dresden Architecture of an efficient MEMS final test system Dr. Martin Brucke, SPEKTRA Dresden Architecture of an efficient MEMS final test system Outline Introduction of SPEKTRA What are typical Measurement tasks

More information

Aerospace & Defense Application Areas and Case Studies

Aerospace & Defense Application Areas and Case Studies Aerospace & Defense Application Areas and Case Studies AVICHAL KULSHRESTHA Technical Marketing Engineer National Instruments Graphical System Design A platform-based approach for measurement and control

More information

FlexRIO. FPGAs Bringing Custom Functionality to Instruments. Ravichandran Raghavan Technical Marketing Engineer. ni.com

FlexRIO. FPGAs Bringing Custom Functionality to Instruments. Ravichandran Raghavan Technical Marketing Engineer. ni.com FlexRIO FPGAs Bringing Custom Functionality to Instruments Ravichandran Raghavan Technical Marketing Engineer Electrical Test Today Acquire, Transfer, Post-Process Paradigm Fixed- Functionality Triggers

More information

Intelligent sonic online flow measurement. Pouriya Niknam Piramoon System Qeshm, Iran

Intelligent sonic online flow measurement. Pouriya Niknam Piramoon System Qeshm, Iran Intelligent sonic online flow measurement Pouriya Niknam Piramoon System Qeshm, Iran Table of Content Introduction measurement principal modeling verification Challenges and conclusion Introduction The

More information

A new architecture for real-time control in RFX-mod G. Manduchi, A. Barbalace Big Physics Symposium 1/16

A new architecture for real-time control in RFX-mod G. Manduchi, A. Barbalace Big Physics Symposium 1/16 A new architecture for real-time control in RFX-mod G. Manduchi, A. Barbalace 2011 Big Physics Symposium 1/16 Current RFX control system MHD mode control Plasma position control Toroidal field control

More information

Thunderbolt 3 PCIe Expansion Chassis with DisplayPort - PCIe x16

Thunderbolt 3 PCIe Expansion Chassis with DisplayPort - PCIe x16 Thunderbolt 3 PCIe Expansion Chassis with DisplayPort - PCIe x16 Product ID: TB31PCIEX16 This Thunderbolt 3 PCIe expansion chassis lets you add a PCI Express card to your laptop or desktop computer, through

More information

Techniques to Reduce Measurement Errors in Challenging Environments

Techniques to Reduce Measurement Errors in Challenging Environments Techniques to Reduce Measurement Errors in Challenging Environments Jon Semancik VTI Instruments Corp. 2031 Main Street Irvine, CA 92614 949-955-1894 jsemancik@vtiinstruments.com INTRODUCTION Test engineers

More information

Engineering Challenges in Developing Large Flash Memory System

Engineering Challenges in Developing Large Flash Memory System Engineering Challenges in Developing Large Flash Memory System August 2010 Presented by Kumar Venkatramani Introduction Scalable Modular Case Study of System Design Scalable and Modular Design Image Storage

More information

LabVIEW control solutions on Mac. Christophe Salzmann. Ecole Polytechnique Fédérale de Lausanne - Switzerland

LabVIEW control solutions on Mac. Christophe Salzmann. Ecole Polytechnique Fédérale de Lausanne - Switzerland 1 LabVIEW control solutions on Mac Christophe Salzmann Ecole Polytechnique Fédérale de Lausanne - Switzerland 2 1! Outline Education & research context Control on the Mac (timeline) Current solutions Research

More information

Analog Input Sample Rate

Analog Input Sample Rate ECONseries Low Cost USB Data Acquisition Modules Overview The ECONseries is a flexible yet economical series of multifunction DAQ modules. You chse the number of analog I/O and digital I/O channels, the

More information

The Concept of Sample Rate. Digitized amplitude and time

The Concept of Sample Rate. Digitized amplitude and time Data Acquisition Basics Data acquisition is the sampling of continuous real world information to generate data that can be manipulated by a computer. Acquired data can be displayed, analyzed, and stored

More information

Solving the Data Transfer Bottleneck in Digitizers

Solving the Data Transfer Bottleneck in Digitizers Solving the Data Transfer Bottleneck in Digitizers With most modern PC based digitizers and data acquisition systems a common problem is caused by the fact that the ADC technology usually runs in advance

More information

X-ray imaging software tools for HPC clusters and the Cloud

X-ray imaging software tools for HPC clusters and the Cloud X-ray imaging software tools for HPC clusters and the Cloud Darren Thompson Application Support Specialist 9 October 2012 IM&T ADVANCED SCIENTIFIC COMPUTING NeAT Remote CT & visualisation project Aim:

More information

An FPGA-based High Speed Parallel Signal Processing System for Adaptive Optics Testbed

An FPGA-based High Speed Parallel Signal Processing System for Adaptive Optics Testbed An FPGA-based High Speed Parallel Signal Processing System for Adaptive Optics Testbed Hong Bong Kim 1 Hanwha Thales. Co., Ltd. Republic of Korea Young Soo Choi and Yu Kyung Yang Agency for Defense Development,

More information

Enhancing VXIbus Systems Through the Power of the Ethernet

Enhancing VXIbus Systems Through the Power of the Ethernet Enhancing VXIbus Systems Through the Power of the Ethernet by Tom Sarfi, VTI Instruments Abstract - The VXIbus architecture has proven to be an ideal solution for high channel count applications that require

More information

New Interconnnects. Moderator: Andy Rudoff, SNIA NVM Programming Technical Work Group and Persistent Memory SW Architect, Intel

New Interconnnects. Moderator: Andy Rudoff, SNIA NVM Programming Technical Work Group and Persistent Memory SW Architect, Intel New Interconnnects Moderator: Andy Rudoff, SNIA NVM Programming Technical Work Group and Persistent Memory SW Architect, Intel CCIX: Seamless Data Movement for Accelerated Applications TM Millind Mittal

More information

ITER Fast Plant System Controller Prototype Based on PXIe Platform

ITER Fast Plant System Controller Prototype Based on PXIe Platform ITER Fast Plant System Controller Prototype Based on PXIe Platform M. Ruiz, J. Vega, R. Castro, J.M. López, E. Barrera, G. Arcas, D. Sanz, J. Nieto, B.Gonçalves, J. Sousa, B. Carvalho, N. Utzel, P. Makijarvi

More information

Microsecond Latency, Real-Time, Multi-Input/Output Control using GPU Processing

Microsecond Latency, Real-Time, Multi-Input/Output Control using GPU Processing Microsecond Latency, Real-Time, Multi-Input/Output Control using GPU Processing Nikolaus Rath March 20th, 2013 N. Rath (Columbia University) µs Latency Control using GPU Processing March 20th, 2013 1 /

More information

LabVIEW FPGA in Hardware-in-the-Loop Simulation Applications

LabVIEW FPGA in Hardware-in-the-Loop Simulation Applications LabVIEW FPGA in Hardware-in-the-Loop Simulation Applications Publish Date: Dec 29, 2008 38 Ratings 4.16 out of 5 Overview Hardware-in-the-loop (HIL) simulation is achieving a highly realistic simulation

More information

End-to-End Adaptive Packet Aggregation for High-Throughput I/O Bus Network Using Ethernet

End-to-End Adaptive Packet Aggregation for High-Throughput I/O Bus Network Using Ethernet Hot Interconnects 2014 End-to-End Adaptive Packet Aggregation for High-Throughput I/O Bus Network Using Ethernet Green Platform Research Laboratories, NEC, Japan J. Suzuki, Y. Hayashi, M. Kan, S. Miyakawa,

More information

A Hardware-Friendly Bilateral Solver for Real-Time Virtual-Reality Video

A Hardware-Friendly Bilateral Solver for Real-Time Virtual-Reality Video A Hardware-Friendly Bilateral Solver for Real-Time Virtual-Reality Video Amrita Mazumdar Armin Alaghi Jonathan T. Barron David Gallup Luis Ceze Mark Oskin Steven M. Seitz University of Washington Google

More information

A variety of ECONseries modules provide economical yet flexible solutions. Waveform Generation

A variety of ECONseries modules provide economical yet flexible solutions. Waveform Generation ECONseries BUS: USB Type: Economy, Mini-Instruments ECONseries Economy USB Mini-Instruments Flexible Yet Economical A variety of low-cost ECONseries modules are available to provide flexible yet economical

More information

High-Speed M Series Multifunction DAQ 16-Bit, up to 1.25 MS/s, up to 80 Analog Inputs

High-Speed M Series Multifunction DAQ 16-Bit, up to 1.25 MS/s, up to 80 Analog Inputs High-Speed M Series Multifunction DAQ 16-Bit, up to 1.25 MS/s, up to 80 Inputs NI M Series High-Speed NI recommends high-accuracy M Series for 5X more measurement sensitivity and lowpass filtering 16,

More information

The Cray CX1 puts massive power and flexibility right where you need it in your workgroup

The Cray CX1 puts massive power and flexibility right where you need it in your workgroup The Cray CX1 puts massive power and flexibility right where you need it in your workgroup Up to 96 cores of Intel 5600 compute power 3D visualization Up to 32TB of storage GPU acceleration Small footprint

More information

We Survey Design and Modeling Framework for Towed Multimeasurement Seismic Streamer Data

We Survey Design and Modeling Framework for Towed Multimeasurement Seismic Streamer Data We-04-12 Survey Design and Modeling Framework for Towed Multimeasurement Seismic Streamer Data K. Eggenberger* (Schlumberger), P. Christie (Schlumberger), M. Vassallo (Schlumberger) & D.J. van Manen (Schlumberger)

More information

DAQ in MATLAB HANS-PETTER HALVORSEN,

DAQ in MATLAB HANS-PETTER HALVORSEN, Telemark University College Department of Electrical Engineering, Information Technology and Cybernetics DAQ in MATLAB HANS-PETTER HALVORSEN, 2011.06.07 Faculty of Technology, Postboks 203, Kjølnes ring

More information