Moving NN Triggers to Level-1 at LHC Rates Triggering Problem in HEP Adopted neural solutions Specifications for Level 1 Triggering Hardware Implementation.

Slides:



Advertisements
Similar presentations
Introduction to Programmable Logic John Coughlan RAL Technology Department Electronics Division.
Advertisements

1 KU College of Engineering Elec 204: Digital Systems Design Lecture 9 Programmable Configurations Read Only Memory (ROM) – –a fixed array of AND gates.
EELE 367 – Logic Design Module 2 – Modern Digital Design Flow Agenda 1.History of Digital Design Approach 2.HDLs 3.Design Abstraction 4.Modern Design Steps.
Graduate Computer Architecture I Lecture 15: Intro to Reconfigurable Devices.
FPGA-Based System Design: Chapter 3 Copyright  2004 Prentice Hall PTR SRAM-based FPGA n SRAM-based LE –Registers in logic elements –LUT-based logic element.
BIST for Logic and Memory Resources in Virtex-4 FPGAs Sachin Dhingra, Daniel Milton, and Charles Stroud Electrical and Computer Engineering Auburn University.
JET Algorithm Attila Hidvégi. Overview FIO scan in crate environment JET Algorithm –Hardware tests (on JEM 0.2) –Results and problems –Ongoing work on.
Characterization Presentation Neural Network Implementation On FPGA Supervisor: Chen Koren Maria Nemets Maxim Zavodchik
Lecture 26: Reconfigurable Computing May 11, 2004 ECE 669 Parallel Computer Architecture Reconfigurable Computing.
Tutorial on Neural Networks Prévotet Jean-Christophe University of Paris VI FRANCE.
ENGIN112 L38: Programmable Logic December 5, 2003 ENGIN 112 Intro to Electrical and Computer Engineering Lecture 38 Programmable Logic.
The Multicycle Processor II CPSC 321 Andreas Klappenecker.
The Spartan 3e FPGA. CS/EE 3710 The Spartan 3e FPGA  What’s inside the chip? How does it implement random logic? What other features can you use?  What.
DSP online algorithms for the ATLAS TileCal Read Out Drivers Cristobal Cuenca Almenar IFIC (University of Valencia-CSIC)
Presenting: Itai Avron Supervisor: Chen Koren Final Presentation Spring 2005 Implementation of Artificial Intelligence System on FPGA.
Evolution of implementation technologies
Programmable logic and FPGA
Presenting: Itai Avron Supervisor: Chen Koren Characterization Presentation Spring 2005 Implementation of Artificial Intelligence System on FPGA.
Final Presentation Neural Network Implementation On FPGA Supervisor: Chen Koren Maria Nemets Maxim Zavodchik
Characterization Presentation Neural Network Implementation On FPGA Supervisor: Chen Koren Maria Nemets Maxim Zavodchik
CMPUT Computer Organization and Architecture II1 CMPUT329 - Fall 2003 Topic: Internal Organization of an FPGA José Nelson Amaral.
Implementing Digital Circuits Lecture L3.1. Implementing Digital Circuits Transistors and Integrated Circuits Transistor-Transistor Logic (TTL) Programmable.
Presenting: Itai Avron Supervisor: Chen Koren Mid Semester Presentation Spring 2005 Implementation of Artificial Intelligence System on FPGA.
CS 151 Digital Systems Design Lecture 38 Programmable Logic.
GallagherP188/MAPLD20041 Accelerating DSP Algorithms Using FPGAs Sean Gallagher DSP Specialist Xilinx Inc.
EKT303/4 PRINCIPLES OF PRINCIPLES OF COMPUTER ARCHITECTURE (PoCA)
Introduction to FPGA AVI SINGH. Prerequisites Digital Circuit Design - Logic Gates, FlipFlops, Counters, Mux-Demux Familiarity with a procedural programming.
A comprehensive method for the evaluation of the sensitivity to SEUs of FPGA-based applications A comprehensive method for the evaluation of the sensitivity.
DLS Digital Controller Tony Dobbing Head of Power Supplies Group.
Electronics in High Energy Physics Introduction to Electronics in HEP Field Programmable Gate Arrays Part 1 based on the lecture of S.Haas.
System Arch 2008 (Fire Tom Wada) /10/9 Field Programmable Gate Array.
Building Asynchronous Circuits With JBits Eric Keller FPL 2001.
Research on Reconfigurable Computing Using Impulse C Carmen Li Shen Mentor: Dr. Russell Duren February 1, 2008.
Neural Network Architectures Aydın Ulaş 02 December 2004
J. Christiansen, CERN - EP/MIC
FPGA (Field Programmable Gate Array): CLBs, Slices, and LUTs Each configurable logic block (CLB) in Spartan-6 FPGAs consists of two slices, arranged side-by-side.
VHDL Project Specification Naser Mohammadzadeh. Schedule  due date: Tir 18 th 2.
FPGA-Based System Design: Chapter 3 Copyright  2004 Prentice Hall PTR FPGA Fabric n Elements of an FPGA fabric –Logic element –Placement –Wiring –I/O.
FPGA-Based System Design: Chapter 3 Copyright  2004 Prentice Hall PTR Topics n FPGA fabric architecture concepts.
Introduction to FPGA Created & Presented By Ali Masoudi For Advanced Digital Communication Lab (ADC-Lab) At Isfahan University Of technology (IUT) Department.
Lecture #3 Page 1 ECE 4110–5110 Digital System Design Lecture #3 Agenda 1.FPGA's 2.Lab Setup Announcements 1.HW#2 assigned Due.
EE3A1 Computer Hardware and Digital Design
Anurag Dwivedi. Basic Block - Gates Gates -> Flip Flops.
Rinoy Pazhekattu. Introduction  Most IPs today are designed using component-based design  Each component is its own IP that can be switched out for.
Algorithms for the ROD DSP of the ATLAS Hadronic Tile Calorimeter
Samuel Silverstein Stockholm University CMM++ firmware development Backplane formats (update) CMM++ firmware.
M.Mohajjel. Why? TTM (Time-to-market) Prototyping Reconfigurable and Custom Computing 2Digital System Design.
ESS | FPGA for Dummies | | Maurizio Donna FPGA for Dummies Basic FPGA architecture.
CDA 4253 FPGA System Design RTL Design Methodology 1 Hao Zheng Comp Sci & Eng USF.
EEL 5722 FPGA Design Fall 2003 Digit-Serial DSP Functions Part I.
FPGA-Based System Design: Chapter 3 Copyright  2004 Prentice Hall PTR Topics n FPGA fabric architecture concepts.
Introduction to Field Programmable Gate Arrays (FPGAs) EDL Spring 2016 Johns Hopkins University Electrical and Computer Engineering March 2, 2016.
Programmable Logic Devices
Hardware Descriptions of Multi-Layer Perceptions with Different Abstraction Levels Paper by E.M. Ortigosa , A. Canas, E.Ros, P.M. Ortigosa, S. Mota , J.
Reconfigurable Architectures
Topics SRAM-based FPGA fabrics: Xilinx. Altera..
Introduction to Programmable Logic
Electronics for Physicists
FPGA Implementation of Multicore AES 128/192/256
Implementation of the Jet Algorithm ATLAS Level-1 Calorimeter Trigger
DESIGN AND IMPLEMENTATION OF DIGITAL FILTER
Field Programmable Gate Array
Field Programmable Gate Array
Field Programmable Gate Array
Final Project presentation
Electronics for Physicists
RTL Design Methodology Transition from Pseudocode & Interface
Implementing Logic Gates and Circuits
Programmable logic and FPGA
Presentation transcript:

Moving NN Triggers to Level-1 at LHC Rates Triggering Problem in HEP Adopted neural solutions Specifications for Level 1 Triggering Hardware Implementation Results Conclusion Proposed architecture Jean-Christophe Prévotet L aboratoire des I nstruments et S ystèmes d’ I le de F rance

Triggering problem in High Energy Physics Detector Level 1 Trigger ~1µs Level 2 Trigger ~20µs Level 3 Trigger Level 4 Trigger Offline event reconstruction Reject Dedicated Hardware Implementation Conventional Microprocessors Incoming data from sub-detectors Y~0 Background Y~1 Physics

Level 1 Trigger Level 2 Trigger Hardware Adopted Solutions Latency of 500ns => No digital circuits possible OR Straightforward Circuits made of RAMs : lack of precision, small networks Latency of 10µs => Possible use of digital circuits Exple: CNAPS in the H1 experiment => 8µs to execute a 64x64x1 net DSPs Current solutions Future solutions Technology trend enables to transpose L2 complexity of neural computations into L1

Level 1 Trigger Scheme Preprocessor Digitization Pre-Sums, … Analog signals from the calorimeter Output data To Level2 (every 25ns) Main control module 500ns Neural processing FPGAsDemultiplex unitMultiplex unit Data arrive each BC (25ns) and processed in a time multiplexed way Timing Specifications of the ATLAS experiment at LHC

Specifications …… Execution time : 500 ns Weights coded in 16 bits States coded in 8 bits with data arriving every BC=25ns Electrons, tau, hadrons, jets

Neural processor Architecture TanH PE I/O module Control unit Matrix of n*m Processing Elements (PEs) 256 PEs for a 128x64x4 network PE TanH 1 matrix row computes a neuron ACC Control unit I/O module The result is back-propagated To calculate the output layer TanH are stored in LUT

PE architecture X AccumulatorMultiplier Weights mem Input data8 16 Addr gen + Data in cmd bus Control Module Data out

Row Accumulator Trunc Registers Din Input bus (data coming from other rows) Output bus (data going to other rows) Adder Multiplexers / Demultiplexers Truncation unit Register bank

Hardware Implementation in a FPGA What is a FPGA… I/O Ports Block Rams Programmable connections Programmable Logic Blocks DLL LUT Carry & Control Carry & Control DQ DQ y yq xb x xq cin cout G4 G3 G2 G1 F4 F3 F2 F1 bx Xilinx Virtex slice

Results Timing Time in clock cycles for the whole neural net : around 60 cycles. Target Clock frequencyProcessing time: 8.33ns120MHz => VIRTEX2 compatible Global synthesis and implementation on the FPGA Timing and resources optimization What has to be done… What is done today… Description of the whole design in VHDL Functionnal simulations of the different modules (Multipliers, acc, control, PE..) Individual Modules synthesis (translated into logic blocks)

Flexibility Implementation in a FPGA => easily re-configurable Processing time: doesn’t really depend on the number of neurons in the hidden layer Coding precision easily changeable Weight Precision, activation functions, etc. Advantages 1 neuron = 4 added PEs Disadvantages Resources consuming => many FPGAs required Summary Implementation of digital neural network feasible in real time Transposition of level2 concepts into Level 1 Proposed architecture Fewer performances than custom circuits