HimalSherchanLogBooK

Table of Contents

Sandbox.LogBook 11/28/2012

• Added a possible fix for the legends problem for the DCS O2O Webpage.
  • Somehow the legend color were flipped in the plots.
  • The data was being accessed asynchronously (function call running in the background while other functions call were being made).
  • The color of the legends is based on whatever data is accessed first.
  • Even though the low voltage data is accessed first, the asynchronous functionality might have led to adding the high voltage data before the low voltage data.
• Worked toward switching the plots to a local system instead of relying on a Google API.

Sandbox.LogBook 1/23/2013

• Finished making final changes to the webpage.
• DCS O2O Webpage:
  • The webpage consists of three regular plots and three overview plots. The overview plots are there as guide for the regular plots.
    • Every single plot has some sort of time seperation(24hrs, 13hrs, and 1hr) between the last plotted point and the edge of the plot.
    • The plot shows the number of modules that are turned on at a certain time.
  • Call to the Google API is made in the beginning. The call was made to the google server to begin with. However for security purposes, it is done locally.
    • Download jquery-1.3.2.js (which can be found through Google)
    • <script type="text/javascript", src="jquery-1.3.2.js"></script>
  • All the files with the prefix "oneMonth" indicate one month's data and the files with the prefix "full" indicate total data accumulated so far since 2008.
    • Data files: full_hv(24hrs delay), full_hv_13hoursDelay (13hrs delay), full_hv_last (1hr delay), full_lv(24hrs delay), full_lv_13hoursDelay (13hrs delay), full_lv_last (1hr delay), oneMonth_hv(24hrs delay), oneMonth_hv_13hoursDelay (13hrs delay), oneMonth_hv_last (1hr delay), oneMonth_lv(24hrs delay), oneMonth_lv_13hoursDelay (13hrs delay), oneMonth_lv_last (1hr delay). "HV" stands for High Voltage. "LV" stands for Low Voltage
    • Variable with qsParm holds the file name. For example: qsParm[data] = 'oneMonth'. The file name is acquired through the url. Everytime one of the plots are changed from full to oneMonth or vice versa, the url is modified. The modified url is parsed to get the file name
  • The most important change added to the webpage was creating a seperation(24hrs, 13hrs, and 1hr) between the last plotted data point and the edge of the plot. This change was helpful for the offline shifters to make sense out of the plot.Without the seperation, it would have seemed like the plot was still storing data.
    • The files are accessed using AJAX request.
    • AJAX calls a success function based on the call made to the url.
    • The data for that given file is accessed through the success function
    • The format of the data in the file is [milliseconds, no. of modules]
    • Javascript uses January 1, 1970 to use as reference for the milliseconds.
    • The AJAX call is made very early so that it can modify the plot.
    • For each plot there is a unique AJAX call as well as success function.
    • Since there are three plots, each success function extracts the last value of their respective data file and adds a certain amount of milliseconds to it.
  • The zoom functionality used for the regular plot is rectangular zooming. The overview plot just uses selection functionality.
    • Different types of zooming were tested. However, the rectangular zooming seemed like the most optimal for the regular plot.
    • The zooming that was based on scrolling was applied. However, the refresh rate was not fast enough for the plot to alter quickly.
    • So the current zoom functionality contains rectangular zooming. Each plot has a zoom out button that takes viewer back to the most zoomed out view of the plot.

Sandbox.LogBook 4/8/2013

Web interface to access the sistrip cabling information: detector, FEDs, FECs and power supply (CMS Tracker Software - CMS Track 16)

• Python program written to parse the SiStripDetCabling object
• The SiStrip Detector consists of sensors (modules), FED and FED channels, FEC, control rings and CCUs and power supplies and power supply channels.
• —The Link between sensors, FEDs and FECs are described by the conditional object known as SiStripDetCabling
• —Dumped as text file in the CMSSW job page
• Sample text file: https://test-stripdbmonitor.web.cern.ch/test-stripdbmonitor/CondDBMonitoring/cms_orcoff_prod/CMS_COND_31X_STRIP/DBTagCollection/SiStripFedCabling/SiStripFedCabling_GR10_v1_hlt/CablingLog/CablingInfo_Run100357.txt
• CMSSW job page: https://its.cern.ch/jira/browse/CMSTRACK-16
• Python program written to parse the SiStripDetectorCabling object
• The text object is huge.
• Search for patters that make it easier to parse
• Used a block of text as a cue for a new channel information
• Program looks for the word "FedCrate" in every block to indicate the start of the block
• Following that, it loops through each block and parse line by line based on certain characters