DQCaloCombTestShifterGettingStarted < Sandbox

%RED% *------------------------------------- LAST UPDATED: April 2018 -------------------------------------* %ENDCOLOR%
<br />%STARTINCLUDE%
%IMAGE{"twikiBanner_gettingStarted.001.png" size="1129"}%

%DASHBOARD{ section="dashboard_start" }%
%DASHBOARD{ section="box_start" title="How to use this page" width="992" height="150" }%
This pages gives an overview of the shift. Click the links to go deeper. <br />

%TOC%

Happy hunting!
%DASHBOARD{ section="box_end" }%
%DASHBOARD{ section="dashboard_end" }%

---+ Shift overview

   * You'll look at some plots in a run and decide if they look abnormal.
   * You'll write a brief report of your findings and save it in the logbook.
   * Then you'll mark _both_ that you've completed checking the run, and that flag any defects.
   * Finally, you'll present a summary of your findings, consolidating defects across runs, in the DQ meeting. The more carefully you document as you go, the easier this will be!

*But why are you doing this?* <br />
I mean, aside from your institute bugging you about OTP, why is this shift structured the way it is?<br /> 
As data is collected, its reconstruction occurs at Tier-0, which is subdivided into multiple streams. <br />
The express stream contains a subset (~10%) of the full data volume; it is reconstructed in quasi-real time, with the aim of obtaining results before the main reconstruction starts.<br />
Currently, we are taking data with a _36 hour_ calibration loop, with, generally, only one long run per day. <br />
There is a first pass of reconstruction on the *Express* stream (or ES1 in the web-display) and, once this has been reviewed, 36 hours later the full stream processing is started at Tier0 including any updates for cells to be masked, beamspot position, etc. <br />
This full-stream processing is labeled as *Bulk* (or BLK in the web-display).<br />

%BLUE% *The duty of the shifter is to look at the first pass processing of the data in order to highlight any problems which can be solved in the bulk processing and then to check the bulk for the runs which have finished the processing to assess whether the problems have indeed been solved.* %ENDCOLOR%

 To clarify:
   * You'll look at each run _twice_: 
      * Firstly, at 10% of the run, in the express stream.  
      * Secondy, at 100% of the run in the bulk stream, which is available ~48 hours after the express stream.
   * Process the express stream the *same day* it becomes available.
   * Aim to process the bulk stream the same day it becomes available. DO NOT leave all of the bulk processing for the next shifter! :)
   
---+ What you should know _before_ your first shift

   * Know your responsibilities, e.g. 
      * which defects should you set flags for?
      * when should you contact the !CaloCombined experts and/or subdetector experts
      * which runs are you meant to process? Which can you leave for the next shifter?
   * Know how to get the information you need:
      * Do you know where to find the plots? The whiteboard? Who's on shift?
   * Know who your expert shifter is, the shifter in the week before you, the shifter in the week after.

---+ Get started!

Details of how the shift works will be explained over the pages below. We suggest going through them in order. <br />
Remember, the first few days of your first shift will likely take a bit if time as there is a lot of information to digest, but by mid-week you'll get into the swing of things. <br />
If in doubt - poke your expert shifter! 

   * *Page 1:* [[DQCaloCombDailyShifterWorkflow][The Daily Shifter Workflow]]
   * *Page 2:* [[DQCaloCombWeeklyShifterWorkflow][The Weekly Shifter Workflow]]
   * *Page 3:* [[DQCaloCombSubDetectorsAndFolders][SubDetectors and Folder strategy]]
   * *Page 4:* [[DQCaloCombDocumentation][Documentation: writing a log, the DB, what to include in your weekly summary]]
   * *Page 5:* [[DQCaloCombWeeklyPlotIndex][Plot Index: Everything you need to know about almost every plot]]


-- Main.RebeccaCarney - 2018-04-01

Attachments

Topic attachments
I	Attachment	History	Action	Size	Date	Who	Comment
png	twikiBanner_gettingStarted.001.png	r1	manage	71.0 K	2018-04-02 - 00:32	RebeccaCarney

Topic revision: r2 - 2018-04-02 - RebeccaCarney

Account
- Log In

- Cern Search
- TWiki Search
- Google Search
Sandbox All webs

Edit
Attach

Copyright &© 2008-2021 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback