Strip cluster

[VinInn]

Two Major component here:

1. Speed up of a factor two of the Strip Clusterizer
2. new infrastructure based on edmNew::DetSetVector for the OnDemand tracking
   this version is fully backward compatible and support both old and new mechanisms
   for reco and hlt.

This development has been discussed in depth within HLT/TSG, Tracker DPG and Tracking POG.
see
https://indico.cern.ch/event/301711/ (same slides in Tracker POG)
and Mia's slide in https://indico.cern.ch/event/300088/

irrelevant regressions can be observed mostly due to changes in the calibration "accuracy" (double to float etc)

Once this PR will be in the release, HLT will port the config to the new on demand mechanism.
in next (pre)-release cleanup of code will start.

It is therefore rather urgent to have an official pre-release incorporating this PR in order not to slow down the work on the HLT side

[cmsbuild]

A new Pull Request was created by @VinInn (Vincenzo Innocente) for CMSSW_7_1_X.

Strip cluster

It involves the following packages:

CalibFormats/SiStripObjects
CalibTracker/SiStripESProducers
CondFormats/SiStripObjects
DQM/SiStripMonitorHardware
DataFormats/Common
DataFormats/SiStripCluster
EventFilter/SiStripRawToDigi
RecoLocalTracker/SiStripClusterizer
RecoLocalTracker/SubCollectionProducers
RecoTracker/MeasurementDet
RecoTracker/TkTrackingRegions

@apfeiffer1, @diguida, @danduggan, @rovere, @cmsbuild, @anton-a, @Dr15Jones, @rcastello, @deguio, @slava77, @ggovi, @Degano, @ojeda, @ktf, @thspeer, @nclopezo can you please review it and eventually sign? Thanks.
@ghellwig, @wmtan, @GiacomoSguazzoni, @rovere, @gpetruc, @cerati this is something you requested to watch as well.
You can sign-off by replying to this message having '+1' in the first line of your reply.
You can reject by replying to this message having '-1' in the first line of your reply.
@nclopezo, @ktf you are the release manager for this.
You can merge this pull request by typing 'merge' in the first line of your comment.

[nclopezo]

@VinInn
Hi Vincenzo,

this pull request is unmergeable, could you please rebase it?

[VinInn]

@rcastello, @diguida
please refer to the second part of the talk at https://indico.cern.ch/event/301711/
for details.

[gennai]

ah ah ah ah !

S.
Il giorno 19/feb/2014, alle ore 14:46, Vincenzo Innocente notifications@github.com ha scritto:

di principio diguida e rcastello dovrebbero saltare dalla sedia…
scommetto che invece approveranno senza fare domande…
(tanto dovrebbe essere tracker dpg a fare domande visto che gli ho cambiato tutto il modo con cui applicano le calibrazioni alle strip, ma tracker dpg non firma)
v.

On 19 Feb, 2014, at 2:42 PM, Giulio Eulisse notifications@github.com wrote:

@deguio, @danduggan, @rcastello, @diguida can you please review this?

—
Reply to this email directly or view it on GitHub.

[diguida]

+1
Thanks @VinInn for pointing to the documentation. For now, we approve, even if I see that the way conditions are consumed in the tracker are completely revisited. For the future, may I ask to keep AlCa in the loop?
Thanks

[ktf]

DQM change is just commenting out an header file. I'll bypass their signature. @davidlange6

[perrotta]

@slava77 @VinInn others

I am preparing possible backports of HLT related code in 70X.

At the ORP yestarday we agreed that (#2378 + #2527 + #2571) can be backported, and I am right now testing the PR I've put up with them.

In order to complete the backport of everything needed to replicate the CMSSW_7_1_0_pre3 HLT template in 70x I'd also need to backport this Pull Request. Before embarking myself in that work, would you agree backporting this PR in 70X (given the comment #2466 (comment) earlier in this thread)?

Thank you,
Andrea

[VinInn]

I do not disagree with the back-port.
I hope that no major merge conflicts are found...

[slava77]

OK, Andrea. Go ahead with bport.
Unless I missed something, there were no technical issues running this code in pre3 relvals.

[diguida]

Hi!
Fine also from our side. Please let us know how you want to deal with validation.