Characteristics of Reconfigurable Hardware

Slides:



Advertisements
Similar presentations
Introduction to Storage Area Network (SAN) Jie Feng Winter 2001.
Advertisements

System Area Network Abhiram Shandilya 12/06/01. Overview Introduction to System Area Networks SAN Design and Examples SAN Applications.
Addressing the System-on-a-Chip Interconnect Woes Through Communication-Based Design N. Vinay Krishnan EE249 Class Presentation.
Reconfigurable Computing: What, Why, and Implications for Design Automation André DeHon and John Wawrzynek June 23, 1999 BRASS Project University of California.
Chapter 13 Embedded Systems
Zheming CSCE715.  A wireless sensor network (WSN) ◦ Spatially distributed sensors to monitor physical or environmental conditions, and to cooperatively.
Seven Minute Madness: Special-Purpose Parallel Architectures Dr. Jason D. Bakos.
I/O Channels I/O devices getting more sophisticated e.g. 3D graphics cards CPU instructs I/O controller to do transfer I/O controller does entire transfer.
HS/DSL Project Yael GrossmanArik Krantz Implementation and Synthesis of a 3-Port PCI- Express Switch Supervisor: Mony Orbach.
Heterogeneous Computing Dr. Jason D. Bakos. Heterogeneous Computing 2 “Traditional” Parallel/Multi-Processing Large-scale parallel platforms: –Individual.
1 A survey on Reconfigurable Computing for Signal Processing Applications Anne Pratoomtong Spring2002.
Module I Overview of Computer Architecture and Organization.
Networking Virtualization Using FPGAs Russell Tessier, Deepak Unnikrishnan, Dong Yin, and Lixin Gao Reconfigurable Computing Group Department of Electrical.
ECE 526 – Network Processing Systems Design Network Processor Architecture and Scalability Chapter 13,14: D. E. Comer.
A brief overview about Distributed Systems Group A4 Chris Sun Bryan Maden Min Fang.
Paper Review: XiSystem - A Reconfigurable Processor and System
Cluster Computers. Introduction Cluster computing –Standard PCs or workstations connected by a fast network –Good price/performance ratio –Exploit existing.
Computer Architecture Lecture10: Input/output devices Piotr Bilski.
To be smart or not to be? Siva Subramanian Polaris R&D Lab, RTP Tal Lavian OPENET Lab, Santa Clara.
ATCA based LLRF system design review DESY Control servers for ATCA based LLRF system Piotr Pucyk - DESY, Warsaw University of Technology Jaroslaw.
Jump to first page One-gigabit Router Oskar E. Bruening and Cemal Akcaba Advisor: Prof. Agarwal.
Disk WP-4 “Information Technology” J. Hogenbirk/M. de Jong  Introduction (‘Antares biased’)  Design considerations  Recent developments  Summary.
The Cosmic Cube Charles L. Seitz Presented By: Jason D. Robey 2 APR 03.
MAPLD 2005/254C. Papachristou 1 Reconfigurable and Evolvable Hardware Fabric Chris Papachristou, Frank Wolff Robert Ewing Electrical Engineering & Computer.
An Architecture and Prototype Implementation for TCP/IP Hardware Support Mirko Benz Dresden University of Technology, Germany TERENA 2001.
L/O/G/O Input Output Chapter 4 CS.216 Computer Architecture and Organization.
PARALLEL PROCESSOR- TAXONOMY. CH18 Parallel Processing {Multi-processor, Multi-computer} Multiple Processor Organizations Symmetric Multiprocessors Cache.
Development of Programmable Architecture for Base-Band Processing S. Leung, A. Postula, Univ. of Queensland, Australia A. Hemani, Royal Institute of Tech.,
VLVnT09A. Belias1 The on-shore DAQ system for a deep-sea neutrino telescope A.Belias NOA-NESTOR.
Axel Jantsch 1 Networks on Chip Axel Jantsch 1 Shashi Kumar 1, Juha-Pekka Soininen 2, Martti Forsell 2, Mikael Millberg 1, Johnny Öberg 1, Kari Tiensurjä.
Survey of multicore architectures Marko Bertogna Scuola Superiore S.Anna, ReTiS Lab, Pisa, Italy.
Cluster Computers. Introduction Cluster computing –Standard PCs or workstations connected by a fast network –Good price/performance ratio –Exploit existing.
1 Device Controller I/O units typically consist of A mechanical component: the device itself An electronic component: the device controller or adapter.
Online Software November 10, 2009 Infrastructure Overview Luciano Orsini, Roland Moser Invited Talk at SuperB ETD-Online Status Review.
Generic and Re-usable Developments for Online Software Slow Control, Configuration, Data Format & Online Processing Shebli Anvar, CEA Irfu January 12,
Lecture 13 Parallel Processing. 2 What is Parallel Computing? Traditionally software has been written for serial computation. Parallel computing is the.
Firmware and Software for the PPM DU S. Anvar, H. Le Provost, Y.Moudden, F. Louis, E.Zonca – CEA Saclay IRFU – Amsterdam/NIKHEF, 2011 March 30.
Computer Networking A Top-Down Approach Featuring the Internet Introduction Jaypee Institute of Information Technology.
WPFL General Meeting, , Nikhef A. Belias1 Shore DAQ system - report on studies A.Belias NOA-NESTOR.
MOTHER BOARD PARTS BY BOGDAN LANGONE BACK PANEL CONNECTORS AND PORTS Back Panels= The back panel is the portion of the motherboard that allows.
Giovanna Lehmann Miotto CERN EP/DT-DI On behalf of the DAQ team
These slides are based on the book:
Use of FPGA for dataflow Filippo Costa ALICE O2 CERN
DAQ read out system Status Report
Lynn Choi School of Electrical Engineering
“FPGA shore station demonstrator for KM3NeT”
Challenges in ALICE and LHCb in LHC Run3
Dynamo: A Runtime Codesign Environment
Enrico Gamberini, Giovanna Lehmann Miotto, Roland Sipos
Chapter 7.2 Computer Architecture
System On Chip.
What is Fibre Channel? What is Fibre Channel? Introduction
CS184a: Computer Architecture (Structure and Organization)
Multi-Processing in High Performance Computer Architecture:
Understanding the OSI Reference Model
GGF15 – Grids and Network Virtualization
Programmable Logic Controllers (PLCs) An Overview.
Anne Pratoomtong ECE734, Spring2002
CS 31006: Computer Networks – The Routers
Architecture of Parallel Computers CSC / ECE 506 Summer 2006 Scalable Programming Models Lecture 11 6/19/2006 Dr Steve Hunter.
Chapter 4: Threads.
Overview of Computer Architecture and Organization
The performance requirements for DSP applications continue to grow and the traditional solutions do not adequately address this new challenge Paradigm.
Overview of Computer Architecture and Organization
Network Processors for a 1 MHz Trigger-DAQ System
Cloud-Enabling Technology
Computer Networking A Top-Down Approach Featuring the Internet
ADSP 21065L.
Cluster Computers.
Reconfigurable Computing (EN2911X, Fall07)
Presentation transcript:

Characteristics of Reconfigurable Hardware Reconfigurable Hardware for on-shore DAQ of km3 scale Neutrino Telescopes Hardware vs. Software Characteristics of Reconfigurable Hardware Opportunities for DAQ Tasos Belias

Application specific Hardware vs. Software fast “local execution” fine-grained parallelism no extraneous connections compact operators tailored to function simple control direct wire connections between operators Software Slow sequential execution overhead time “interpreting” operations Inefficient Area fixed width operators, may not match problem general operators, bigger than required area to store instructions, control execution Fast But Fixed! But Flexible! Combine both through Reconfigurable Hardware Tasos Belias

Reconfigurable Hardware FPGA-Advantage Customization of operator type, width, and interconnect of I/O data. Exploit low overhead and parallelism of application. Exploit cases where operation can be bound and then reused a large number of times. Adaptation to any switch-fabric protocol for data routing to external resources. Tasos Belias

n-Telescope on-shore DAQ tasks Key to the project: simultaneous, dead time-less readout of O(10000) channels and signal processing tasks integrity check of raw data, apply calibrations to raw data, classification algorithms using raw data, as well as network and protocol layer processing for distribution / routing of raw data to processing nodes. To achieve these tasks: use a system that provides both traditional and reconfigurable hardware Scalability Flexibility Components Off The Shelf-Implementations Tasos Belias

KM3NeT targeted Implementations Accumulate deep-sea data on dedicated links Physical medium interface Optical / Electrical converter, Small form-factor pluggable transceivers Protocol layer MAC core, Custom link Check and Filter Data FPGA-centric memory & processing Standard FPGA interconnects ( such as Rocket I/O) Concurrent, adaptable filtering schemes Physics-interest filters, GRBs, SNs,… Format and Route to processing farms Adapt to any standard switch-fabric protocol Tasos Belias

Possible Mixed Architecture PCIe I/O Scaling 21GB/s (peak) System Memory FSB / FPGA North Bridge 10GB/s PCIe South Bridge FSB 8.5GB/s FSB / FPGA 10GB/s Conventional processor centric MP host Host for Power / Control Route downstream Data Monitoring & Visualization Reconfigurable data centric system Data Acquisition Check, Calibrate, Classify Real-Time & Deterministic Tasos Belias 6

All on one slide Reconfigurable architectures - FPGAs computing in hardware programmable like software more computation per unit area than processors efficient where processors are inefficient Heterogeneous architectures (mix processors, reconfigurable) “general-purpose” and “application-targeted” processing components Exploiting these architectures  new strategies for DAQ of km3 scale n-Telescope Tasos Belias