R&D for First Level Farm Hardware Processors Joachim Gläß Computer Engineering, University of Mannheim Contents –Overview of Processing Architecture –Requirements.

Slides:



Advertisements
Similar presentations
Hao wang and Jyh-Charn (Steve) Liu
Advertisements

Digital RF Stabilization System Based on MicroTCA Technology - Libera LLRF Robert Černe May 2010, RT10, Lisboa
Programmable Logic Devices
Presents The Silver Family An Integrated Approach to Processors, Data Communication and Head End Integration.
An ATCA and FPGA-Based Data Processing Unit for PANDA Experiment H.XU, Z.-A. LIU,Q.WANG, D.JIN, Inst. High Energy Physics, Beijing, W. Kühn, J. Lang, S.
CHEP04 - Interlaken - Sep. 27th - Oct. 1st 2004T. M. Steinbeck for the Alice Collaboration1/20 New Experiences with the ALICE High Level Trigger Data Transport.
Programmable logic and FPGA
Final Presentation Neural Network Implementation On FPGA Supervisor: Chen Koren Maria Nemets Maxim Zavodchik
XUP Virtex-5 Development System January XUP Virtex52 Introducing XUPV5-LX110T A powerful and versatile platform packaged and priced for Academia!
1 Design of the Front End Readout Board for TORCH Detector 10, June 2010.
Lecture 7 Lecture 7: Hardware/Software Systems on the XUP Board ECE 412: Microcomputer Laboratory.
System Architecture A Reconfigurable and Programmable Gigabit Network Interface Card Jeff Shafer, Hyong-Youb Kim, Paul Willmann, Dr. Scott Rixner Rice.
PCI/104 Explanation and Uses in Test Program Set Development.
General FPGA Architecture Field Programmable Gate Array.
© Copyright Xilinx 2004 All Rights Reserved 9 November, 2004 XUP Virtex-II Pro Development System.
NetBurner MOD 5282 Network Development Kit MCF 5282 Integrated ColdFire 32 bit Microcontoller 2 DB-9 connectors for serial I/O supports: RS-232, RS-485,
Christian Steinle, University of Mannheim, Institute of Computer Engineering1 L1 Tracking – Status CBMROOT And Realisation Christian Steinle, Andreas Kugel,
DLS Digital Controller Tony Dobbing Head of Power Supplies Group.
Uni-Heidelberg, KIP, V.Angelov 1 International Workshop TRDs – Present & Future September, Romania Wafer Tester, Optical Link, GTU V. Angelov Kirchhoff.
RiceNIC: A Reconfigurable and Programmable Gigabit Network Interface Card Jeff Shafer, Dr. Scott Rixner Rice Computer Architecture:
LECC2003 AmsterdamMatthias Müller A RobIn Prototype for a PCI-Bus based Atlas Readout-System B. Gorini, M. Joos, J. Petersen (CERN, Geneva) A. Kugel, R.
Research on Reconfigurable Computing Using Impulse C Carmen Li Shen Mentor: Dr. Russell Duren February 1, 2008.
NEDA collaboration meeting at IFIC Valencia, 3rd-5th November 2010 M. Tripon EXOGAM2 project Digital instrumentation of the EXOGAM detector EXOGAM2 - Overview.
Understanding Data Acquisition System for N- XYTER.
PHENIX upgrade DAQ Status/ HBD FEM experience (so far) The thoughts on the PHENIX DAQ upgrade –Slow download HBD test experience so far –GTM –FEM readout.
J. Christiansen, CERN - EP/MIC
L1Topo-phase0 Uli Schäfer 1. Topo GOLD successfully used to explore technologies and initially verify 6.4Gb/s link integrity over moderate length electrical.
StreamBlade TM StreamBlade TM Applications Rev 1.2.
TELL1 The DAQ interface board for LHCb experiment Gong guanghua, Gong hui, Hou lei DEP, Tsinghua Univ. Guido Haefeli EPFL, Lausanne Real Time ,
Status of Global Trigger Global Muon Trigger Sept 2001 Vienna CMS-group presented by A.Taurok.
CPT Week, April 2001Darin Acosta1 Status of the Next Generation CSC Track-Finder D.Acosta University of Florida.
Frank Lemke DPG Frühjahrstagung 2010 Time synchronization and measurements of a hierarchical DAQ network DPG Conference Bonn 2010 Session: HK 70.3 University.
CCD Cameras with USB2.0 & Gigabit interfaces for the Pi of The Sky Project Grzegorz Kasprowicz Piotr Sitek PERG In cooperation with Soltan Institute.
Summary Computing and DAQ Walter F.J. Müller, GSI, Darmstadt 5 th CBM Collaboration Meeting GSI, March 9-12, 2005.
Data Acquisition Backbone Core J. Adamczewski-Musch, N. Kurz, S. Linev GSI, Experiment Electronics, Data processing group.
1 CS/COE0447 Computer Organization & Assembly Language CHAPTER 1.
FEE Electronics progress PCB layout progress VHDL progress in TBU Prototype fixtures FEE64 commissioning A few of the remaining tasks 16th July 2009.
1 Aerospace Data Storage and Processing Systems SEAKR Engineering Proprietary Information SEAKR Engineering Inc. On-Board Processing SEAKR Engineering.
Design Criteria and Proposal for a CBM Trigger/DAQ Hardware Prototype Joachim Gläß Computer Engineering, University of Mannheim Contents –Requirements.
Lecture 12: Reconfigurable Systems II October 20, 2004 ECE 697F Reconfigurable Computing Lecture 12 Reconfigurable Systems II: Exploring Programmable Systems.
Latest ideas in DAQ development for LHC B. Gorini - CERN 1.
Fast Tracking of Strip and MAPS Detectors Joachim Gläß Computer Engineering, University of Mannheim Target application is trigger  1. do it fast  2.
1 SysCore for N-XYTER Status Report Talk by Dirk Gottschalk Kirchhoff Institut für Physik Universität Heidelberg.
XLV INTERNATIONAL WINTER MEETING ON NUCLEAR PHYSICS Tiago Pérez II Physikalisches Institut For the PANDA collaboration FPGA Compute node for the PANDA.
D. Attié, P. Baron, D. Calvet, P. Colas, C. Coquelet, E. Delagnes, R. Joannes, A. Le Coguie, S. Lhenoret, I. Mandjavidze, M. Riallot, E. Zonca TPC Electronics:
17/02/06H-RORCKIP HeidelbergTorsten Alt The new H-RORC H-RORC.
LECC2004 BostonMatthias Müller The final design of the ATLAS Trigger/DAQ Readout-Buffer Input (ROBIN) Device B. Gorini, M. Joos, J. Petersen, S. Stancu,
Project number: Andreas Sakellariou – PRISMA ELECTRONICS 1 st EU Mid-Term Review meeting 7 November 2014.
KM3NeT Offshore Readout System On Chip A highly integrated system using FPGA COTS S. Anvar, H. Le Provost, F. Louis, B.Vallage – CEA Saclay IRFU – Amsterdam/NIKHEF,
Cloudland Instruments Hawkeye Electronics Snapshot March 10th, 2016.
Digital Logic & Design Dr.Waseem Ikram Lecture No. 43.
Mu3e Data Acquisition Ideas Dirk Wiedner July /5/20121Dirk Wiedner Mu3e meeting Zurich.
PC-based L0TP Status Report “on behalf of the Ferrara L0TP Group” Ilaria Neri University of Ferrara and INFN - Italy Ferrara, September 02, 2014.
DOM Electronics (Digital Optical Module) 1 WPFLElectronics PPMDOM ElectronicsF. Louis.
PRM for AM06 Daniel Magalotti Collaboration between: KIT, INFN Pisa and INFN Perugia.
The trigger-less readout for the Mu3e experiment Dirk Wiedner On behalf of the Mu3e collaboration 31 March 20161Dirk Wiedner.
29/05/09A. Salamon – TDAQ WG - CERN1 LKr calorimeter L0 trigger V. Bonaiuto, L. Cesaroni, A. Fucci, A. Salamon, G. Salina, F. Sargeni.
Eric Hazen1 Ethernet Readout With: E. Kearns, J. Raaf, S.X. Wu, others... Eric Hazen Boston University.
Future Hardware Development for discussion with JLU Giessen
TELL1 A common data acquisition board for LHCb
The Train Builder Data Acquisition System for the European-XFEL
CoBo - Different Boundaries & Different Options of
COVER Full production should arrive today
Electronics for Physicists
Development of new CN for PXD DAQ
Table 1: The specification of the PSICM and the ePSICM Prototypes
Characteristics of Reconfigurable Hardware
Electronics for Physicists
TELL1 A common data acquisition board for LHCb
Programmable logic and FPGA
Presentation transcript:

R&D for First Level Farm Hardware Processors Joachim Gläß Computer Engineering, University of Mannheim Contents –Overview of Processing Architecture –Requirements for Hardware Processors –R&D for Hardware Processors –Conclusion March 10, 2005 CBM Collaboration Meeting

Overview of Processing Architecture Processing resources Hardware processors –L1/FPGA Software processors –L1/CPU Active Buffers Sub-farm network –Pnet Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Overview of Processing Architecture Processing resources Sub-farm size –1000 x 10 Gbit/s detector output links –storing and sorting according to epochs/events –several epochs per sub- farm (no duplicate MAPS data) one scenario –64 sub-farms –16 epochs stored and processed per subfarm –4 AB units (á 4 ABs) –8 L1/FPGA units (á 4 processors) –32 L1/CPU Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Overview of Processing Architecture Processing resources Active Buffer (AB) –interface between detector link, Bnet and Pnet store data from detector link and output to Bnet receive sorted data according to epochs from Bnet event building according to timestamps within one epoch output event data to processing units –histogramming of timestamps, data transfer protocols, control of large memory => FPGA Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Overview of Processing Architecture Processing resources Hardware processors (L1/FPGA) –process 10 Gbit/s input data stream in real-time –most reasonable technology: FPGA => L1/FPGA –processing in parallel hardware + flexibility of programming –forefront of chip development (regular internal structure) > 100 k logic gates, > 500 MHz for complex algorithms integrated multi gigabit transmitter (MGT): up to 20 x 10 Gbit/s serial I/O Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Overview of Processing Architecture Processing resources Software processors (L1/CPU) –general purpose processors PCs system-on-a-chip processors (better GFlops/€ or GFlops/Watt ?) –IBM Blue Gene, STI Cell processor, Strech S5000 Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Overview of Processing Architecture Processing resources Sub-farm network (Pnet) –connections only inside one sub-farm (short distances ~1 m) –freely programmable although for fixed algorithm only fixed point-to-point links NO all-to-all network, like Bnet –usage of built-in MGTs of FPGAs COTS switches PCIe plausible candidate Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Requirements for Hardware Processors Algorithms in FPGA functional:-> simulation timing:-> simulation, (R&D prototype) External memories (data buffers, large memories, …) –e.g. LUTs for Hough transform: several MByte (ca. 20 addresses) => ZBT SRAM functional:-> simulation timing:-> simulation, R&D prototype (clock feedback, termination, …) –e.g. Active Buffer: several tenth of MBytes (1000 x 10µs x 1GByte/s) => DDR SDRAM functional:-> simulation timing:-> simulation, R&D prototype (clock feedback, termination, …) Interconnection network –e.g. Pnet, MGTs up to 10 Gbit/s –test physical layer of multi gigabit communication chip-to-chip:-> simulation of impedances, R&D prototype board-to-board: –optical:-> chip-to-chip –backplane:-> simulation of impedances, connectors, R&D prototype Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Architecture of R&D Prototype communication via backplane –4 boards, all-to-all –different length of traces –up to 10 Gbit/s serial –=> FR4 Rogers DDR ZBT FPGA connector SFPSFP XC2VPX20 Flash RS232 Ethernet PPC zeroXT 10GB SMT Ethernet Flash DDR Linux µC FPGA with MGTs –up to 10 Gbit/s serial –=> XC2VPX20 (8 x MGT) –=> XC2VPX70 (20 x MGT) externals –2 x ZBT SRAM –2 x DDR SDRAM –for PPC: Flash, Ethernet, … initialization and control –standalone board/system –microcontroller running Linux Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering

Conclusion R&D prototype to learn: –physical layer of communication 2.5 Gbit/s up to 10 Gbit/s chip-to-chip board-to-board (-> connectors, backplane) PCB layout, impedances PCB material (FR4, Rogers, …) –next step: communication protocols –more resources needed => XC2VPX70?, Virtex4? (availability?) –external memories fast controllers for ZBT and DDR RAM PCB layout, termination, … Joachim Gläß, Univ. Mannheim, Institute of Computer Engineering