12 March 2009. Technology and Instrumentation in Particle Physics 2009, Tsukuba, Japan.
PPTX,
PPT,
PDF
The Compact Muon Solenoid (CMS) experiment at the LHC incorporates a 66 million channel silicon pixel detector at its center for track seeding and precise vertexing. The hardware and data acquisition software of the pixel detector are briefly introduced in this paper. Experience with installation is succinctly recounted and some of the calibrations performed during the commissioning period are described. Preliminary performance numbers like charge collection and hit residuals from cosmic ray tracks are presented.
Proceedings paper: 
2 September 2008. Cornell University. A walk through the transportation, installation and commissioning of the pixel detector. The commissioning is split into phases I, II and III over which various tests and calibrations were performed, leading up to the detection of cosmic muons leaving tracks through our detector. Presented at the Cornell University CMS meeting on the 2nd of September 2008.
16 July 2008. CERN. Lessons learnt for Pixel Online Software from CRUZET 3. Problems resulting from the integration of the Detector Power Up Procedure with Run Control and Monitoring System addressed. PixelTKFECSupervisor going into indefinite waits when reading the DCU temperatures mentioned. Problems with shipping data to the PixelFEDSpySupervisor for the Beam Radiation Monitoring group highlighted along with peripheral issues with visibility of new terabyte disks.
9 July 2008. CERN. A graphic description of how the Detector Power Up Procedure is implemented with POS-DCS Integration at the levels of PixelSupervisor, PixelTKFECSupervisor and PixelFECSupervisor is presented.
25 June 2008. CERN. A first report of a working Detector Power Up Procedure with Pixel Online Software and Detector Control System integration.
19 June 2008. CERN. A report on our first attempt to run a calibration through the central DAQ.
18 June 2008. CERN. The close-to-final plan for Pixel Online Software - Detector Control System Integration required for the detector startup presented after extensive discussions with the Detector Control System expert.
18 June 2008. CERN. A report on the SLink test with the central DAQ of the 8 Forward Pixel FEDs. The trigger throttling system could not be tested.
11 June 2008. CERN. A presentation on the readiness of Pixel Online Software for CRUZET 2. Includes the installation of Run Control and Monitoring System, XDAQ, Pixel Online Software and the nuances of their deployment.
4 June 2008. CERN. A comprehensive explanation of how to turn on and off the booking of official run numbers from PixelSupervisor presented. The problems of un-configured FEDs and the disappearance of log messages raised at CRUZET 1 solved.
22 May 2008. CERN. Problems from CRUZET 1 mentioned in the 14 May 2008 presentation are addressed. The problem of run number collisions is solved.
14 May 2008. CERN. Lessons learnt from global running at CRUZET 1 for Pixel Online Software. The problems of run number collisions, un-configured FEDs at the end of runs, disappearance of log messages, inadequate monitoring of the FED FIFOs and limited disk space are mentioned.
17 April 2008. CERN. Talk at Pixel DAQ Tutorial. A cursory introdution to the hierarchical state machine structure of Pixel Online Software for Pixel Shifters.
9 April 2008. CERN. Introduced the new Configuring state in the FSM of PixelSupervisor, PixelFECSupervisor, and PixelFECSupervisor. This was done to 1. Parallelize the configuration of FEC and FED Supervisors, 2. Wait for DCS system to ramp up voltage with hanging up the DAQ system, 3. To be able to report percentage completion to the GUI during configuration.
8 April 2008. CERN. Talk at the Tracker Detector Performance Group Meeting outlining the construction, software preparation and participation with the Pixels-In-a-Box at CMS Global Run March (GRUMM) 2008.
19 March 2008. CERN. Recounting the first experience of participating in a CMS Global Run at a Cornell CMS Meeting.
12 March 2008. CERN. CMS Pixels participates successfully at CMS March 2008 Global Run with the Pixels-In-a-Box! Pixel Online Software worked seamlessly with Central Run Control, Global Trigger and the Event Builder Farm. Several calibrations were carried out with Pixel Online Software and several runs of data taken with triggers from the Muon Drift Tubes. My log of these runs as the Pixel Shifter is located here. Pixels data from Tier 0 analyzed by Pixel Offline Software.
20 February 2008. CERN. Final steps towards XDAQ - DCS Integration stalled by problems with DCS software. All other aspects tested.
13 February 2008. CERN. The problem with FED 11 at Point 5 pinned down and requesting replacement. Trigger rates > 500 Hz cause throttling.
13 February 2008. CERN. Further progress with XDAQ - DAQ Integration reported.
30 January 2008. CERN. The steps required for the DAQ and Detector Control System integration outlined. Initial progress reported.
19 December 2007. CERN. At Point 5, emulated FED data is processed by the central DAQ. Triggers go missing. CRC errors observed on FED ID 11. TTCci clock noticed to not lock.
19 December 2007. CERN. New Pixel Alive, S-Curve and Gain Calibrations that use the SLink introduced. A TK-FEC Low Level GUI is begun.
30 November 2007. CERN. Talk at the Technical Board Meeting on readiness for participation in a Global Run at Point 5.
29 November 2007. Cornell University. Talk at Cornell University CMS Meeting on Pixel Online Software's integration with Global Run Control and further plans at Point 5.
28 November 2007. CERN. Brief report of outstanding FED readout problems.
21 November 2007. CERN. Presentation of the Online Software preparation for the CMS November Global Run and the SLink test. Ready for both. Several problems regarding the TTS signals got solved.
7 November 2007. CERN. A step-through presentation of the software spawning and control hierarchy of Pixel Online Software installed at Point 5 and how it fits in with the CMS Global DAQ. Presented at the Online Software Commissioning Meeting.
2 November 2007. CERN. Presentation at the Technical Board Meeting outlining current status at Point 5 and level of preparation for November Global Run.
31 October 2007. CERN. Status report of our preparation for participation in November Global Run: TTS Test passed, Local SLink Test passed. Yet to try Global SLink test with Central DAQ group.
26 October 2007. CERN. Results from our TTS Test with the CMS Global DAQ reported at the Pixel Technical Board Meeting.
24 October 2007. CERN. A new Finite State Machine for Pixel Online Software proposed that can decouple calibration code from the GUI, thus allowing one to stop or pause the calibration without crashing XDAQ and report completeion status on the GUI.
10 October 2007. CERN. A clear outline of the steps required to participate in the CMS Global Runs.
30 September 2007. CERN. A presentation of the AJAX structure used for the virtual oscilloscopes on the PixelFEDSupervisor's Low Level GUIs.
21 September 2007. CERN. Outline of the Finite State Machine for an SLink monitoring XDAQ application written for use at the Tracker Integration Facility.
12 November 2007. CERN. First steps towards installing Pixel Online Software at Point 5 (CMS Site) taken.
15 August 2007. CERN. New TTS states added to FSM in PixelFunctionManager, PixelSupervisor and PixelFEDSupervisor.
8 August 2007. CERN. PixelFEDSupervisor can write more than 2 GB per file. Low Level GUI of PixelSupervisorGUI transferred to PixelSupervisor. Usage of Run Numbers introduced.
1 August 2007. CERN. Run Control can now completely control the XDAQ part of Pixel Online Software. PixelSupervisorGUI abandoned. GUI functionalities given to PixelSupervisor.
18 July 2007. CERN. Run Control can spawn our XDAQ Online Software but problems with configuring it persists.
11 July 2007. CERN. Expandable HTML/Javscript summary tables for FED Baseline and Address Level Calibrations available in PixelFEDSupervisor now. Miscellaneous problems with other calibration procedures reported. The RunType associated with the "Configure" command from Run Control is identified with our Alias.
27 June 2007. CERN. XDAQ Supervisors organised into zones and groups. SOAP messages from PixelFunctionManager to PixelSupervisor possible.
18 June 2007. CERN. Presentation of PixelFunctionManager status at CMS Week.
2 May 2007. CERN. Error Handling system contributed by Stefan Spanier's group integrated with PixelFEDSupervisor. Spawning of new thread in PixelFEDSupervisor for collecting physics data successful but occassional oddities in the data seen and reported.
25 April 2007. CERN. Starting to take real pixel hits data using a radioactive source using Pixel Online Software! Bugs with FEC's ROC level GUI made.
18 April 2007. CERN. More progress with Run Control and Monitoring System and XDAQ Online Software integration. FEC's ROC level GUI allows one to change mask and trim settings for each pixel and save configurations to file. Internal changes to PixelSupervisor.
5 April 2007. CERN. Low level hardware problems regarding missing triggers, FED Baseline Calibration and Cal-Delay shifts reported and solved with Mauro Dinardo.
4 April 2007. CERN. PixelFECSupervisor's drill-down GUI enhanced with AJAX so the browser does not need to refresh. Realtime control over the FEC elements is now possible.
28 March 2007. CERN. Partial integration of Pixel Online Software with CMS Run Control databases achieved. Several bugs in FED Baseline and Address Level calibrations fixed. A drill-down GUI for the PixelFECSupervisor introduced which allows users to change settings at the FEC, module and ROC levels.
14 March 2007. CERN. An outline of what the obstacles are in integrating existing Pixel Online Software with CMS Run Control and a plan to overcome them.
23 February 2007. CERN. A Technical Board Meeting Presentation showing off the overall structure and progress of Pixel Online Software till then.
21 February 2007. CERN. Finite State Machine changes implemented for the "Stop" transition and for the Trigger Throttling System tests.
15 February 2007. Cornell University. A brief presentation at the CU CMS Meeting announcing release of Pixel Online Software Version 1.1.0. Outlines the structure and capabilities of Pixel Online Software Version 1.1.0. Mentions the features under development for Version 1.2.0
7 February 2007. Cornell University. A first version of Pixel Online Software Manual drafted. Sliders, TTS signals and more functionality added to FED Low Level GUI.
24 January 2007. Cornell University. Extending Version 1.1.0 to accommodate multiple FED crates and FEDs per crate. Low Level GUIs changed.
14 January 2007. Johns Hopkins University. Talk at Pixel Online Software Workshop detailing Baseline and Address Level Calibrations
11 January 2007. Johns Hopkins University. Talk at Pixel Online Software Workshop outlining the SOAP messaging and State Machine structures of current Pixel Online Software.
11 January 2007. Johns Hopkins University. Talk at Pixel Online Software Workshop detailing the structure and function of each Supervisor.
9 November 2006. Fermilab. Gain Calibration Curve taken. Pixel Detector successfully configured from files instead of the database.
1 November 2006. Fermilab. Baseline and Address Level Calibrations demonstrated. Data for Clock Phase and Delay Calibration can be taken but selection done by eye!
10 August 2006. Cornell University. Pixel FEC and FED Supervisors being given their independent GUIs.
3 August 2006. Fermilab. Pixel FED Supervisor successfully tested with hardware at Fermilab. Bugs found in Pixel FEC Supervisor and Pixel FEC Interface.
27 July 2006. Cornell University. Shipped FED data to RU. Expandable PixelSupervisorGUI.
29 June 2006. Fermilab, Cornell University. The Pixel Supervisor XDAQ application successfully triggered detection of hits using the Trigger and Timing Control system. The Front End Driver extracted this data and shipped it to the Read Out Unit Builder for assimilation.
22 June 2006. Fermilab, Cornell University. Channel Offset Calibration and Address Level Calibration routines were developed, successfully integrated into the Pixel Suite and tested against hardware at Fermilab.
24 May 2006. Fermilab, Cornell University. The simulation of fake data at the Front End Driver in the absence of the Read Out Chip was made more user friendly.
11 May 2006. Fermilab, Cornell University. The Graphical User Interface for Pixel Supervisor was re-written. The Front End Driver was controlled and tested at Fermilab using this GUI.
27 April 2006. PSI, Vanderbilt University, Fermilab, Cornell University. Will Johns and Danek Kotlinski wrote a first version of XDAQ Libraries for hardware access to the Front End Driver. Hardware calls were integrated into the Pixel Suite and the Front End Driver at Fermilab was operated from the GUI of Pixel Supervisor. Detection of fake pixel hits within the Front End Driver was tested using a VME triggering mechanism.
13 April 2006. Cornell University. A prototype for the Pixel Function Manager RCMS Application is in place. MySQL and Apache Tomcat successfully overlaid with XDAQ. A prototype for the Pixel Front End Driver XDAQ Application is ready.
30 March 2006. Cornell University. XDAQ, our C++ based platform for data aquisition systems was installed. A prototype for the Pixel Supervisor XDAQ Application was readied. A dummy Gain Calibration Algorithm was integrated into it that does not work with hardware. A prototype for the Pixel Front End Controller Supervisor XDAQ Application was readied. A prototype for the Pixel Local Trigger Control Supervisor XDAQ Application was readied.