online overview
DESCRIPTION
Online Overview. L. Coney – UCR MICE CM35 – Feb 2013. Also in This Session:. Controls & Monitoring – Pierrick DAQ – Yordan Online MAUS – Alex Richards MICE Computing – Chris Rogers Will leave the specifics to them.. Note: Online focus here – Operations focus tomorrow. - PowerPoint PPT PresentationTRANSCRIPT
1
Online OverviewOnline Overview
L. Coney – UCRL. Coney – UCR
MICE CM35 – Feb 2013MICE CM35 – Feb 2013
2Coney - CM35 - Feb 2013
Also in This Session:Also in This Session: Controls & Monitoring – PierrickControls & Monitoring – Pierrick
DAQ – Yordan DAQ – Yordan
Online MAUS – Alex RichardsOnline MAUS – Alex Richards
MICE Computing – Chris RogersMICE Computing – Chris Rogers
Will leave the specifics to them..Will leave the specifics to them..
Note: Online focus here – Operations Note: Online focus here – Operations focus tomorrowfocus tomorrow
3Coney - CM35 - Feb 2013
Since CM34 in OctoberSince CM34 in October Significant events in the 4 months Significant events in the 4 months
since the October CM:since the October CM: December runDecember run Christmas breakChristmas break – Failure of AC in – Failure of AC in
PPD computing area – downtime of PPD computing area – downtime of PPD-hosted MICE PPD-hosted MICE services/computingservices/computing
Spectrometer Solenoid Controls Spectrometer Solenoid Controls ReviewReview
Restart of SS2 cooldown/testingRestart of SS2 cooldown/testing Activation run – Wednesday (13 Feb)Activation run – Wednesday (13 Feb)
4Coney - CM35 - Feb 2013
December RunDecember Run Did not go smoothlyDid not go smoothly
Similar to October run issuesSimilar to October run issues Problems with DAQProblems with DAQ
Worked initially but problems developedWorked initially but problems developed Unable to solve remotelyUnable to solve remotely
Problems with C&MProblems with C&M HV control applicationsHV control applications Run ControlRun Control
Problems with Online Monitoring – goneProblems with Online Monitoring – gone
Prompted evaluation of reliability within Prompted evaluation of reliability within Online SystemsOnline Systems Need develop more robust pre-run procedures for Need develop more robust pre-run procedures for
DAQ, C&M, Online ReconstructionDAQ, C&M, Online Reconstruction Need DAQ pulser triggerNeed DAQ pulser trigger Need higher priority on documentation Need higher priority on documentation
More than one person to solve problems – must do More than one person to solve problems – must do better on handover of informationbetter on handover of information
5Coney - CM35 - Feb 2013
December RunDecember Run Need fake data (more than cosmics) – test Need fake data (more than cosmics) – test
full chain of DAQ, unpacker, Online Recofull chain of DAQ, unpacker, Online Reco fake signals from TOFs? – no LED system ‘yet’fake signals from TOFs? – no LED system ‘yet’ Tracker will have LEDsTracker will have LEDs
Emphasized the need for additional Emphasized the need for additional expertiseexpertise
Personnel changes:Personnel changes: New network village manager – Chris Brew New network village manager – Chris Brew
(RAL)(RAL) New RAL network liaison – Antony Wilson (RAL)New RAL network liaison – Antony Wilson (RAL) New DAQ deputy – David Adey (FNAL)New DAQ deputy – David Adey (FNAL) New Online Monitoring owner – Rhys Gardner New Online Monitoring owner – Rhys Gardner
(Brunel grad student)(Brunel grad student) New C&M deputy – ???????????New C&M deputy – ???????????
6Coney - CM35 - Feb 2013
PPD OutagePPD Outage PPD-hosted computing services loss over PPD-hosted computing services loss over
holidayholiday Loss of access to configuration database (CDB) Loss of access to configuration database (CDB)
prevented software development by US MICEprevented software development by US MICE Loss of access to micemineLoss of access to micemine
Highlighted confusion regarding these services Highlighted confusion regarding these services and how they relate to MICE activitiesand how they relate to MICE activities Including Online Group, Data-taking, FC testing, SS Including Online Group, Data-taking, FC testing, SS
testing testing
Prompted review of service loss – see Chris’ talkPrompted review of service loss – see Chris’ talk
Motivated improvement in computing Motivated improvement in computing documentation across the board – in PPD, on documentation across the board – in PPD, on micenet, hardware and servicesmicenet, hardware and services Micenet: Micenet:
http://micewww.pp.rl.ac.uk/projects/computing-software/wihttp://micewww.pp.rl.ac.uk/projects/computing-software/wiki/Micenet_Computerski/Micenet_Computers
PPD: PPD: http://micewww.pp.rl.ac.uk/projects/computing-software/wihttp://micewww.pp.rl.ac.uk/projects/computing-software/wiki/Computing_infrastructureki/Computing_infrastructure
7Coney - CM35 - Feb 2013
Since CM34 cont’dSince CM34 cont’d Spectrometer Solenoid Controls ReviewSpectrometer Solenoid Controls Review
See Pierrick’s talkSee Pierrick’s talk Good feedback Good feedback improvements to the system improvements to the system Led to significant changes in priorities in Led to significant changes in priorities in
C&MC&M Knock-on effect on non-SS C&M workKnock-on effect on non-SS C&M work
Restart of SS2 magnet trainingRestart of SS2 magnet training See Pierrick’s talkSee Pierrick’s talk
Activation RunActivation Run Even with beam only to DSA – still an Even with beam only to DSA – still an
exercise of Online Systems (DAQ, C&M, exercise of Online Systems (DAQ, C&M, Online Reco)Online Reco)
Went much better than December RunWent much better than December Run
8Coney - CM35 - Feb 2013
Overall OnlineOverall Online CompletedCompleted
Automated operating system updates including a Automated operating system updates including a MOM-accessible OFF switch for data-taking MOM-accessible OFF switch for data-taking (stability/performance)(stability/performance)
Spare hardware now organized in R9 (reliability)Spare hardware now organized in R9 (reliability) Computing documentation agreed go on micemine Computing documentation agreed go on micemine
(ease of operations/record keeping)(ease of operations/record keeping) Installation of additional UPSs (reliability)Installation of additional UPSs (reliability)
In progressIn progress Finalize monitoring of new UPS units Finalize monitoring of new UPS units
(reliability/stability)(reliability/stability) Installation of new Online Reconstruction machines Installation of new Online Reconstruction machines
(reliability/stability/performance)(reliability/stability/performance)
NOTE: Infrastructure largely in hand – NOTE: Infrastructure largely in hand – consistently good effort by Matt and Antonyconsistently good effort by Matt and Antony
9Coney - CM35 - Feb 2013
Overall OnlineOverall Online22
DelayedDelayed Installation of new iocpc1 – 21 Dec 2012 (related to Installation of new iocpc1 – 21 Dec 2012 (related to
C&M reliability)C&M reliability) Delayed – requires Pierrick to be at RAL and to Delayed – requires Pierrick to be at RAL and to
coordinate with Matt – last visit December Run coordinate with Matt – last visit December Run Pierrick priority for trip was work on SS controlsPierrick priority for trip was work on SS controls Ran out of timeRan out of time
Computer monitoring info into Alarm Handler – 1 Computer monitoring info into Alarm Handler – 1 Feb 2013 (stability)Feb 2013 (stability)
Requires PierrickRequires Pierrick Pierrick priority is SS controlsPierrick priority is SS controls
Restriction of access to micenet – 1 Feb 2013Restriction of access to micenet – 1 Feb 2013
The link between Online Group, Software The link between Online Group, Software Group, and overall Computing has been weakGroup, and overall Computing has been weak Confusion regarding services, ownership, etc.Confusion regarding services, ownership, etc. Recent work to strengthen this areaRecent work to strengthen this area
10Coney - CM35 - Feb 2013
Online Systems Online Systems Other Other ComputingComputing
MICE must be able to MICE must be able to take data without take data without connection to services or connection to services or computing external to computing external to micenetmicenet
StrengthsStrengths Use primary CDB – Use primary CDB –
located in MLCRlocated in MLCR Can store data in MLCR Can store data in MLCR
for days (so far none for days (so far none deleted from local deleted from local storage although is all storage although is all on GRID)on GRID)
Use local EPICS, Alarm Use local EPICS, Alarm Handler, and Archiver Handler, and Archiver for C&Mfor C&M
DAQ localDAQ local Online Reco uses local Online Reco uses local
MAUS installationMAUS installation
WeaknessesWeaknesses Elog – hosted on PPD – Elog – hosted on PPD –
useful but not criticaluseful but not critical Micemine – hosted on Micemine – hosted on
PPD – run plans, PPD – run plans, documentation, etc. – documentation, etc. – useful but not criticaluseful but not critical
Data Data GRID (see above) GRID (see above) External expert access – External expert access –
Archiver, EPICS gateway, Archiver, EPICS gateway, mousehole – convenience mousehole – convenience rather than necessityrather than necessity
Conclusions Conclusions Current arrangement of Current arrangement of
services works wellservices works well MICE data-taking not at MICE data-taking not at
riskrisk
11Coney - CM35 - Feb 2013
Online ActivitiesOnline Activities Resurrected remote readout of neutron monitor Resurrected remote readout of neutron monitor
– Ian – Ian Updated/improved Online Reco plotsUpdated/improved Online Reco plots
TOFs – DurgaTOFs – Durga CKOVS – Gene CKOVS – Gene Online software – AlexOnline software – Alex
Major developments in C&MMajor developments in C&M Fully developed Focus Coil testingFully developed Focus Coil testing Micenet to R9 (thanks Antony!) Micenet to R9 (thanks Antony!) running as if in MLCR running as if in MLCR Transparent move (from Online Group perspective) to Transparent move (from Online Group perspective) to
MICE HallMICE Hall C&M focused on Spectrometer Solenoid C&M focused on Spectrometer Solenoid
changes in prioritieschanges in priorities C&M milestonesC&M milestones
Implement SS state machine – 1 Feb – see Pierrick’s Implement SS state machine – 1 Feb – see Pierrick’s talktalk
Full SS2 C&M – 1 Feb – see Pierrick’s talkFull SS2 C&M – 1 Feb – see Pierrick’s talk
12Coney - CM35 - Feb 2013
C&M Milestones ShiftingC&M Milestones Shifting Complete Run Control for Step I elements – 21 Dec Complete Run Control for Step I elements – 21 Dec
1 March1 March Depends on data-taking – aimed at completion during December Depends on data-taking – aimed at completion during December
RunRun Depends on Pierrick available – busy on Spectrometer Solenoid Depends on Pierrick available – busy on Spectrometer Solenoid
workwork
Rack room environment monitoring plan – 1 Jan Rack room environment monitoring plan – 1 Jan 1 1 Feb Feb ? ? Require Pierrick – Pierrick busy on SS controlsRequire Pierrick – Pierrick busy on SS controls
Complete HV control user manual – 2 Feb Complete HV control user manual – 2 Feb 1 May 1 May Requires Pierrick to write documentation – Pierrick busy on SS Requires Pierrick to write documentation – Pierrick busy on SS
controls – documentation suffers – trickledown effect on controls – documentation suffers – trickledown effect on OperationsOperations
Complete Run Control manual and shifter guide – 1 Complete Run Control manual and shifter guide – 1 Jan Jan 15 April 15 April Requires Pierrick to write documentation – Pierrick busy on SS Requires Pierrick to write documentation – Pierrick busy on SS
controls – documentation suffers – again, affects Operationscontrols – documentation suffers – again, affects Operations
13Coney - CM35 - Feb 2013
C&M Milestones cont’dC&M Milestones cont’d Bench test pneumatic proton absorber controls – 1 Bench test pneumatic proton absorber controls – 1
April April 1 May 1 May Requires Pierrick – busy on SS controlsRequires Pierrick – busy on SS controls
Install pneumatic proton absorber controls – 15 Install pneumatic proton absorber controls – 15 April April maybe 1 June maybe 1 June Requires Pierrick – busyRequires Pierrick – busy Requires ISIS shutdown: April 1-28 or June 17-30Requires ISIS shutdown: April 1-28 or June 17-30
Implement DS state-based Alarm Handler – 1 Feb Implement DS state-based Alarm Handler – 1 Feb 15 June15 June Priority for first state-based system shifted from DS to SSPriority for first state-based system shifted from DS to SS
New HV control – 15 Jan New HV control – 15 Jan 1 May 1 May Requires Pierrick & slow communication with CAEN – Pierrick Requires Pierrick & slow communication with CAEN – Pierrick
busybusy FC C&M review – 15 FebFC C&M review – 15 Feb
Depends on Pierrick availability and travel capability – not Depends on Pierrick availability and travel capability – not happening todayhappening today
WE NEED ANOTHER C&M PERSON!WE NEED ANOTHER C&M PERSON!
14Coney - CM35 - Feb 2013
Online RecoOnline Reco Improved TOF & CKOV online plots – 21 Dec Improved TOF & CKOV online plots – 21 Dec
2012 – DONE2012 – DONE Requires data-taking with beam to completely vet new Requires data-taking with beam to completely vet new
plots – therefore tied into December Runplots – therefore tied into December Run Install new online reco machines – 1 Feb 2013Install new online reco machines – 1 Feb 2013
Likely to finish roughly on time – required new order by Likely to finish roughly on time – required new order by UniGeneve – installation now during Yordan’s MOM termUniGeneve – installation now during Yordan’s MOM term
KL online plots – 15 April 2013 KL online plots – 15 April 2013 1 June 2013 1 June 2013 No real online plot participation by KL groupNo real online plot participation by KL group Anticipate possible KL plots available after PID paper Anticipate possible KL plots available after PID paper
done?done? EMR online plots – 15 April 2013 EMR online plots – 15 April 2013 1 June 2013 1 June 2013
EMR schedule slipped – online plots follow hardware EMR schedule slipped – online plots follow hardware completioncompletion
NOTE: KL online plots is a guess – no NOTE: KL online plots is a guess – no participation = no plots = no infoparticipation = no plots = no info
15Coney - CM35 - Feb 2013
By June CM36…By June CM36… Finish SS2 and SS1Finish SS2 and SS1
fully developed C&Mfully developed C&M SS state machine SS state machine
completedcompleted
Finish FC – move to HallFinish FC – move to Hall
Documentation – needs Documentation – needs higher priorityhigher priority
Improve stability & Improve stability & reliabilityreliability Complete installation of Complete installation of
iocpc1iocpc1 Complete computer Complete computer
monitoringmonitoring Stabilize C&M applicationsStabilize C&M applications Formalize development vs Formalize development vs
production C&Mproduction C&M Create comprehensive pre-Create comprehensive pre-
run test plan for DAQ and run test plan for DAQ and C&MC&M
SimplifySimplify Production version Run Production version Run
ControlControl Better limits w/in Alarm Better limits w/in Alarm
HandlerHandler Automate datamoverAutomate datamover
Improve functionalityImprove functionality Incorporate EMR into Incorporate EMR into
DAQDAQ Beam test EMRBeam test EMR Bring back Online Bring back Online
MonitoringMonitoring EMR online plotsEMR online plots Online analysis Online analysis
prototypeprototype
New C&M deputy.New C&M deputy.
16Coney - CM35 - Feb 2013
Then….MICE Step IVThen….MICE Step IV Both Spectrometer Solenoids – both Both Spectrometer Solenoids – both
Trackers – one FC and an Absorber module.. Trackers – one FC and an Absorber module.. Requires robust Online systems – DAQ, Requires robust Online systems – DAQ,
C&M, Online Reco & Analysis, C&M, Online Reco & Analysis, InfrastructureInfrastructure
17Coney - CM35 - Feb 2013