Takeo Higuchi IPNS, KEK COPPER Revision and New CPU.

Slides:



Advertisements
Similar presentations
COPPER Status T.Higuchi / KEK Jul.4 th, B2GM Meeting.
Advertisements

1 iTOP Electronics Effort LYNN WOOD PACIFIC NORTHWEST NATIONAL LABORATORY JULY 17, 2013.
A Gigabit Ethernet Link Source Card Robert E. Blair, John W. Dawson, Gary Drake, David J. Francis*, William N. Haberichter, James L. Schlereth Argonne.
12 GeV Trigger Workshop Session II - DAQ System July 8th, 2009 – Christopher Newport Univ. David Abbott.
Takeo Higuchi Institute of Particle and Nuclear Studies, KEK; On behalf of COPPER working group Jan.20, 2004 Hawaii, USA Super B Factory Workshop Readout.
Trigger-less and reconfigurable data acquisition system for J-PET
Takeo Higuchi Institute of Particle and Nuclear Studies, KEK on behalf of the COPPER working group Apr.22,2005Super B Factory Workshop Detector (DAQ/Computing)
1 MICE/AFE II t Update IIT Analog Front End (AFE) Test Stand AFE Conceptual Design Report (CDR) Terry Hart, Illinois Institute of Technology, February.
CALICE – 12/07/07 – Rémi CORNAT (LPC) 1 ASU and standalone test setup for ECAL MAIA BEE project Overview DAQ dedicated Sensor test In situ debug and maintenance.
5 Feb 2002Alternative Ideas for the CALICE Backend System 1 Alternative Ideas for the CALICE Back-End System Matthew Warren and Gordon Crone University.
GEM Electronic System Status New releases of electronic boards: ◦ MPD v 4.0 ◦ Backplane 5 slot short (for UVa design) ◦ APV Front-End with Panasonic connector.
Saverio Minutoli INFN Genova 1 T1 Electronic status Electronic items involved: Anode Front End Card Cathode Front End Card Read-Out Control card Slow Control.
DDL hardware, DATE training1 Detector Data Link (DDL) DDL hardware Csaba SOOS.
September 22, 2005 ESF Workshop-Perugia 1 Virgo Control Electronic upgrade Annecy/Pisa/EGO B.Mours.
TRIGGER-LESS AND RECONFIGURABLE DATA ACQUISITION SYSTEM FOR POSITRON EMISSION TOMOGRAPHY Grzegorz Korcyl 2013.
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.
VIRGO Control System Upgrade: Multi-DSP Board Alberto Gennai INFN Pisa LIGO-G Z.
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.
Detectors and read-out DetectorNo. of ch.Read-out method Device candidates CsI(Tl)768Flash ADCCOPPER + 65 MHz FADC FINESSE Active Polarimeter600 x 12 x.
Takeo Higuchi IPNS, KEK Belle DAQ group Detector R&D: Belle DAQ System 2008/12/06New Hadrons with Various Flavors.
Design and Performance of a PCI Interface with four 2 Gbit/s Serial Optical Links Stefan Haas, Markus Joos CERN Wieslaw Iwanski Henryk Niewodnicznski Institute.
Micro-Research Finland Oy Components for Integrating Device Controllers for Fast Orbit Feedback Jukka Pietarinen EPICS Collaboration Meeting Knoxville.
David Abbott - JLAB DAQ group Embedded-Linux Readout Controllers (Hardware Evaluation)
A PCI Card for Readout in High Energy Physics Experiments Michele Floris 1,2, Gianluca Usai 1,2, Davide Marras 2, André David IEEE Nuclear Science.
Gueorgui ANTCHEVPrague 3-7 September The TOTEM Front End Driver, its Components and Applications in the TOTEM Experiment G. Antchev a, b, P. Aspell.
EUDRB: the data reduction board of the EUDET pixel telescope Lorenzo Chiarelli, Angelo Cotta Ramusino, Livio Piemontese, Davide Spazian Università & INFN.
Performance of the AMT-3 Based TDC System at Belle S.Y.Suzuki, T.Higuchi, Y.Arai, K.Tauchi, M.Nakao, R.Itoh (KEK) H.Nakayama (University of Tokyo)
06/09/2011TIPP Introduction Observational study of supernova explosion (SN) is important for understanding the mechanism of SN in detail. – Observation.
Xiangming Sun1PXL Sensor and RDO review – 06/23/2010 STAR XIANGMING SUN LAWRENCE BERKELEY NATIONAL LAB Firmware and Software Architecture for PIXEL L.
Serial Data Link on Advanced TCA Back Plane M. Nomachi and S. Ajimura Osaka University, Japan CAMAC – FASTBUS – VME / Compact PCI What ’ s next?
Status of shaper prototype production B.G. Cheon (Hanyang U)‏ Dec. KEK 1 st open meeting of the Super KEKB Collaboration.
SuperBelle Event Building System Takeo Higuchi (IPNS, KEK) S.Y.Suzuki (CC, KEK) 2008/12/12 1st Open Meeting of the SuperKEKB Collaboration.
L3 DAQ Doug Chapin for the L3DAQ group DAQShifters Meeting 10 Sep 2002 Overview of L3 DAQ uMon l3xqt l3xmon.
Takeo Higuchi (KEK); CHEP pptx High Speed Data Receiver Card for Future Upgrade of Belle II DAQ 1.Introduction – Belle II DAQ Experimental apparatus.
Sep. 17, 2002BESIII Review Meeting BESIII DAQ System BESIII Review Meeting IHEP · Beijing · China Sep , 2002.
June 17th, 2002Gustaaf Brooijmans - All Experimenter's Meeting 1 DØ DAQ Status June 17th, 2002 S. Snyder (BNL), D. Chapin, M. Clements, D. Cutts, S. Mattingly.
Development of PCI Bus Based DAQ Platform for Higher Luminosity Experiments T.Higuchi, 1 H.Fujii, 1 M.Ikeno, 1 Y.Igarashi, 1 E.Inoue, 1 R.Itoh, 1 H.Kodama,
1 MICE Tracker Readout Update Introduction/Overview TriP-t hardware tests AFE IIt firmware development VLSB firmware development Hardware progress Summary.
Guirao - Frascati 2002Read-out of high-speed S-LINK data via a buffered PCI card 1 Read-out of High Speed S-LINK Data Via a Buffered PCI Card A. Guirao.
Bob Hirosky L2  eta Review 26-APR-01 L2  eta Introduction L2  etas – a stable source of processing power for DØ Level2 Goals: Commercial (replaceable)
Management of the LHCb DAQ Network Guoming Liu *†, Niko Neufeld * * CERN, Switzerland † University of Ferrara, Italy.
Amsterdam, Oct A. Cotta Ramusino, INFN Ferrara 1 EUDRB: status report and plans for interfacing to the IPHC’s M26 Summary: EUDRB developments.
3 Sep 2009SLM1 of 12 SLM performance and limitations based on HW tests.
B. Hirosky 12/14/00 FPGA + FIFO replaces: DMA P/IO buffers TSI device Keep ECL drivers BUY THIS! Same Basic Concept as L2Alpha, but with simplified implementation.
LECC2004 BostonMatthias Müller The final design of the ATLAS Trigger/DAQ Readout-Buffer Input (ROBIN) Device B. Gorini, M. Joos, J. Petersen, S. Stancu,
Rutherford Appleton Laboratory September 1999Fifth Workshop on Electronics for LHC Presented by S. Quinton.
DAQ 1000 Tonko Ljubicic, Mike LeVine, Bob Scheetz, John Hammond, Danny Padrazo, Fred Bieser, Jeff Landgraf.
Status of the SVD DAQ Koji Hara (KEK) 2012/1/16 TRG/DAQ meeting1.
Vladimir Zhulanov for BelleII ECL group Budker INP, Novosibirsk INSTR2014, Novosibirsk 2014/02/28 1.
ROD Activities at Dresden Andreas Glatte, Andreas Meyer, Andy Kielburg-Jeka, Arno Straessner LAr Electronics Upgrade Meeting – LAr Week September 2009.
Super BigBite DAQ & Trigger Jens-Ole Hansen Hall A Collaboration Meeting 16 December 2009.
The Evaluation Tool for the LHCb Event Builder Network Upgrade Guoming Liu, Niko Neufeld CERN, Switzerland 18 th Real-Time Conference June 13, 2012.
Status Report of the PC-Based PXD-DAQ Option Takeo Higuchi (KEK) 1Sep.25,2010PXD-DAQ Workshop.
SVD FADC Status Markus Friedl (HEPHY Vienna) Wetzlar SVD-PXD Meeting, 5 February 2013.
The ALICE Data-Acquisition Read-out Receiver Card C. Soós et al. (for the ALICE collaboration) LECC September 2004, Boston.
E. Hazen1 MicroTCA for HCAL and CMS Review / Status E. Hazen - Boston University for the CMS Collaboration.
Eric Hazen1 Ethernet Readout With: E. Kearns, J. Raaf, S.X. Wu, others... Eric Hazen Boston University.
PXD DAQ (PC option) Status Report
ATLAS Pre-Production ROD Status SCT Version
14-BIT Custom ADC Board Rev. B
Production Firmware - status Components TOTFED - status
Discussion Items Clock PC and/or VME CPU Trigger readout
Iwaki System Readout Board User’s Guide
HCAL Data Concentrator Production Status
Read-out of High Speed S-LINK Data Via a Buffered PCI Card
Evolution of S-LINK to PCI interfaces
PCI BASED READ-OUT RECEIVER CARD IN THE ALICE DAQ SYSTEM
Example of DAQ Trigger issues for the SoLID experiment
The CMS Tracking Readout and Front End Driver Testing
TELL1 A common data acquisition board for LHCb
Presentation transcript:

Takeo Higuchi IPNS, KEK COPPER Revision and New CPU

Part I: Introduction

Readout Electronics Block Diagram PMC CPU local bus PCI bus from detector mezzanine (add-on) modules Network IF Trigger Module Detector I/F Bridge PMC modules FIFO to event builder

“COPPER-II” Digitizer module x 4 Add-on modules for the COPPER RadiSys EPC-6315 – Intel P3 800 MHz – 256 MB memory – Network boot – RedHat Linux 9 Online CPU module VME9U 100BaseT port x 2 Data link & control link Trigger distribution module

COPPER-II Performance Review RX Digitizer Emulator CLK/TRG distribution CPU Ethernet COPPER COPPER-II works w/ L1 rate >30 kHz 10% Data compr. Required L1 rate Typical L1 416 bytes/ev/FINESSE Input L1 rate [kHz] Accepted L1 rate [kHz]

COPPER History in the Belle DAQ – Design started. – Prototype of the COPPER – Upgrade to the COPPER-II. – Proved to work in severer L1 rate (~30 kHz) than the SuperBelle – Debug on AMT3 FINESSE. – EFC DAQ replaced with the COPPER-II’s (6). – Compatibility study btw the COPPER-II and the LeCroy – Part of CDC DAQ replaced with the COPPER-II’s. – More compatibility study. – Deadtime study – Full CDC DAQ replaced with the COPPER-II’s (89). – ACC DAQ replaced with the COPPER-II’s (24) – TRG DAQ replaced with the COPPER-II’s (26). – Study to replace KLM DAQ is going on. Ready to replace all LeCroy DAQ

Deadtime Reduction by COPPER # of hits/TDC DAQ deadtime (μs) Typical data size ~ 29.5 μs ~ 2.8 μs COPPER-II/AMT3 LeCroy By the replacement of the LeCroy with the COPPER-II, DAQ deadtime is reduced by 90%. S.Y.Suzuki

Part II: COPPER Revision

COPPER-II  “COPPER-3” Some parts started to discontinue  Need their replacement with up to date parts. Major motivation of the upgrade. As the technology evolves, price of some parts w/ the higher performance gets much lower: e.g. GbE controller. Fix some minor inconveniences: e.g. move/change switch positions and shapes for better access. Delete unused functions so far.

COPPER-II  COPPER-3 The most important rule of the upgrade: The COPPER-3 shall be fully compatible with the COPPER-II. – FINESSEs, device drivers, and readout software for the COPPER-II shall be used for the COPPER-3 without any modification at all.

List of Major Upgrade Items 1.Replace discontinued parts. 2.Replace discontinuing or out of date parts. –RoHS compliant parts. –Cyclone  Cyclone3. 3.Replace 100Base-T Ethernet controller (82559) w/ GbE controller (82541). 4.Normalize signal line lengths from the TTRX to each of the 4 FINESSEs. –SCLK, TRG etc. 5.Move VME signal receivers to much closer position to the VME J0 connector for more stable signal handling.

List of Major Upgrade Items 6.Change reset switch shape more accessible. –Present switch on the front panel is difficult to push. 7.Move VME base address config. switch to more accessible area. 8.Remove and simplify unused connectors/patterns originally intended for debug use. 9.Join 2 FIFOs to 1. Series of two FIFO chips were aligned to form a larger single FIFO so that we could select FIFO size at production. 10.Change front panel design. FIFO control FIFO control data

COPPER-3 Short Term Schedule Jun. 2 nd :Design work started by the company. Jun. 30 th :Design and schematic chart drawing – delayed. Jul.4 th :----- TODAY Jul.18 th :FPGA programming. Jul.28 th :Pattern layout. Aug.8 th :Board production. Aug.29 th :Parts assembly. Sep.6 th :Board verification. Sep.16 th :Document writing. Sep.26 th :Delivery to KEK. Dec.31 st :End of COPPER-3 study by KEK.

COPPER-3 Long Term Schedule By JFY 2008 – Find out and fix most bugs in the COPPER-3 prototype.  Makes COPPER-3 ready for the mass production. In JFY 2009 – Purchase ~20 COPPER-3.  Single-crate system test. In JFY – Purchase COPPER-3.  Build up sBelle DAQ system. In JFY 2012 – Start sBelle Operation. JFY : Japanese Fiscal Year; from Apr. to Mar.

Part III: New CPU

Roles of the COPPER CPU Major roles related to the data handling: – Readout data from 4 pipeline FIFOs on the COPPER. – Combine an event chunk from each of the FIFOs to form a single event record. – Provide online data monitoring. – Format the event record into the Belle standard. – Transfer the event record to an external readout PC of the COPPER via Ethernet. Peripheral roles: – Initialize COPPER itself, FINESSE, and TTRX. – Handles run control commands: “RUN-START” / “RUN-STOP” / “FATAL-CONDITION”, etc. – Etc.

RadiSys EPC-6315 EPC-6315 is regarded the standard CPU for the COPPER. Why EPC-6315? – RadiSys EPC-6315 was the only commercially-available PrPMC (processor PMC) equipped with the Intel CPU when we started the COPPER R&D. Specs: – CPU: Intel PentiumIII 800 MHz – Memory: 256 MB. – The RJ-45 connector for the Ethernet. – RedHat Linux 7.3 or 9, or FedoraCore 1 Linux run on it. – Price: ~1,100 USD / module. Bootable from CF card or from network.

Cons of the EPC-6315 The EPC-6315 is getting an out-of-date product; its prospect is not very promising. Chipset (RadiSys 82600) deteriorates the data transfer. – PLX9054  PMC memory ­ 112 Mbps (w/ DMA) – PLX9054  EPC-6315 memory ­ 74 Mbps (w/ DMA) Also, we found that RadiSys was less professional in the customer communication… EPC-6315 block diagram The breaks data transfer every 32 bytes in the burst mode, which is the origin of the bottleneck. It cannot be disabled. PMC connector PCI RadiSys mem.

Other Possibilities of PrPMCs GE Fanuc PSL09 – new commercially available PrPMC – CPU: Intel PentiumM 1.4 GHz. – Memory: 256 MB. – Price: ~2,000 USD / module. – 10 COPPER-II’s w/ PSL09 attached had been installed into the DAQ system and have been tested in the beam operation. – It is observed that they worked quite well. It can be one of the possible candidate of the next PrPMC. – I fairy have to say, we haven’t examined if this PrPMC’s chipset would be a bottleneck of the data transfer or not. Photo of PSL09 w/ heat sink removed M.Nakao

Other Possibilities of PrPMCs Possibility of a custom CPU – One Japanese company (Advanet) shows a strong interest to develop a new PrPMC. – According to their rough idea… ­ CPU: Intel Tolapai (code name) 600 MHz equivalent to PentiumM 900MHz. ­ Memory: 512MB-1GB. ­ R&D cost: ~100k USD. ­ Board price: ~1300 USD. ­ R&D will be finished within this fiscal year. We keep surveying any kind of better PrPMC. Looks a bit costly. We are to negotiate with them to find out the possible compromise.

Summary The COPPER-II is to be upgraded to COPPER-3, which has 100% backward compatibility. Within this fiscal year, we will make it ready toward the mass production. Because of some problems, we are searching for the next PrPMC with which the RadiSys’s EPC-6315 is replaced. Good performance of the GE Fanuc’s PSL09 is proved in the beam operation; the PSL09 is one of the candidates. Beside that, we are searching for possibility of the custom PrPMC.

Backup Slides

Readout Electronics Overview Online processor – Data size reduction. – Data link management to event builder. from detector to event builder Online processor Readout FIFO Detector I/F – Signal digitization. – L1 pipeline. Readout FIFO – Event buffers for asynchronous readout.

Design Concept Modules – Detector I/F ­ User defined cards for analog part only, minimizing the R&D cost. – Data reduction CPU ­ Commercially available PMC module, no hardware development cost. Common platform – DAQ software can be generalized. Low Cost Design

Profile of CPU Usage User time: ~2% System time: ~20% Idle time: ~78% = CPU power that is equivalent to P3 ~600MHz is still available Large idle time fraction indicates the PCI bus works at the full performance. – 416 bytes / ADC-module / ev × 40 kHz × 4 ADC modules = bytes/ev/ADC-module

Performance Degradation by Network 416 bytes/ev/ADC-module CPU FIFO Read FIFO Read FIFO Read FIFO Read Network Transfer proc.A proc.B RX Ethernet 11MB/s CPU user time: ~2% CPU system time: ~20% CPU idle time: ~78% Maximum accepted trigger rate: 40 kHz CPU user time: ~2% CPU system time: ~20% CPU idle time: ~78% Maximum accepted trigger rate: 40 kHz CPU user time: ~5% CPU system time: ~31% CPU idle time: ~64% Maximum accepted trigger rate: 32 kHz CPU user time: ~5% CPU system time: ~31% CPU idle time: ~64% Maximum accepted trigger rate: 32 kHz Still works well.