[ ESO ]

Workflow information system for QC

 

WISQ
Other ...
   VLTI/PIONIER
        [page auto-refreshes after 300 sec]   [stop | on]            [press Ctrl+Shift+R to enforce refresh of scores, dates and news]
Workflow information system for QC  
Last update: 2021-04-25T17:50:31 (UT)
  | now: 2024-05-02T19:33:15 (UT)  
General news: NOTE: The Health Check and calChecker services are moved to qcFlow on an instrument by instrument basis. The current pages will not be updated any longer once the move has been finished.
Report news:
png: Information and research:
- png file - advanced studies: QC1 browser | QC1 plotter
Plot
?
Symb
?
Source
*
Average ? Thresholds ? N_
data
QC1
parameter
Data
downloads
Remarks
method value unit method value
1 QC1DB MEDIAN 38.2 min VAL0,60 75 delivery_time this | last_yr | all PIONIER: delays between file acquisition and processing [min];
1a) daytime only, ABs triggering HC updates (broken line: 60min cadence of autoDaily)
1 oQC1DB none   min none  0 delivery_time this | last_yr | all 1b) same, ABs not triggering HC updates

3 QC1DB none   min VAL60,1440 262 delivery_time this | last_yr | all 3a) full time range (upper line: 24 hr); nighttime ABs, failed
3 QC1DB none   min none  1800 delivery_time this | last_yr | all 3b) nighttime ABs, success

4 QC1DB none   min VAL60,1440 0 delivery_time this | last_yr | all 4a) daytime ABs, failed
4 QC1DB none   min none  75 delivery_time this | last_yr | all 4b) daytime ABs, success
*Data sources: QC1DB: QC1 database; LOCAL: local data source
Plot 1
data source:delay_time
(QC1 database)
dataset:
(numbers below apply to this dataset)
delivery_time
median:38.2min
fixed thresholds:0...60min
N_data plotted:75
[click plot for closeup]
Plot 3
data source:delay_time
(QC1 database)
dataset:
(numbers below apply to this dataset)
delivery_time
average:none
fixed thresholds:60...1440min
N_data plotted:262
[click plot for closeup]
Plot 4
data source:delay_time
(QC1 database)
dataset:
(numbers below apply to this dataset)
delivery_time
average:none
fixed thresholds:60...1440min
N_data plotted:0
[click plot for closeup]
This plot

This plot visualizes the delivery times for the autoDaily processing. Delivery time is defined as the time elapsed between the acquisition of the last input raw file in an AB (as recorded in the DATE timestamp in the header) and the finishing of the processing job. The plots distinguish between ABs triggering HC plot updates and other ABs (#1), and between day and nighttime calibrations (#3 and #4). The delivery times are dominated by:
- the cronjob pattern of autoDaily (once per hour),
- the average delay by the DTS,
- the template effect (processing of an AB is possible only after the last input raw frame has been taken and is available from the archive),
- the last-template effect (the current last template is processed only if it is older than one hour, in order to avoid broken templates processing),
- the AB processing time itself,
- possible other effects (the most important one being dependencies).

There are also the 'delay times' stored in the database table, which are defined as time elapsed between the archiving of the first input raw file in an AB (as recorded in the ARCFILE name) and the finishing of the processing job.

The purpose of this plot is to monitor these figure of merit and control the quality of the autoDaily processing scheme. The distinction between day and night calibrations is based on the ARCFILE timestamp: anything between T22 and T09 is called nighttime. Calibration-wise, this time window is normally spent with standard stars and twilight flats only. In exceptional cases daytime calibrations might start before T10, then being counted as nighttime. Disregarding such pollution effects, one can expect that the nighttime calibrations have their own processing pattern which differs strongly from the daytime pattern. Most prominent effect is that they often depend on follow-up daytime calibrations. This is why their delay may sum up to many hours. Therefore they are marked and plotted separately from the daytime calibrations (plot #3).

The plot #1 distinguishes (daytime) calibrations with and without HC update.

The QC1 database table permits filtering by raw_types and other parameters.  


General information

Click on any of the plots to see a close-up version.

The latest date is indicated on top of the plot.

If configured,

  • data points belonging to the latest date are specially marked
  • statistical averages are indicated by a solid line, and thresholds by broken lines
  • outliers are marked by a red asterisk. They are defined as data points outside the threshold lines
  • "aliens" (= data points outside the plot Y limits) are marked by a red arrow ( or )
  • you can download the data for each parameter set if the 'Data downloads' link shows up