InforShell
Trend View

Accessing large files by sampling
(for files over the spreadsheet column or row limit).



Table of Contents

  • Overview
  • How to open files
  • Automatic file size detection
  • Automatic initial sampling



  • Custom re-sampling for more detailed data
  • Getting full resolution data (do not skip records from file)
  • File access with sampling is Read Only



  • Overview

    Some data files are too large to fit in a spreadsheet. E.g. oscilloscope captures may contain 10,000,000 rows. Trend View can enable quick analysis of these files in the spreadsheet by incremental sampling. This is where not all records are loaded in, but for every N rows (or columns), one is read into the spreadsheet. Thus you can analyze data at low resolution, and specify a region of interest. Trend View will load this region again from file at higher sampling resolution. This process can be repeated as many times as necessary, until the sampling interval becomes one.

    Return to Index



    1. Opening large files

  • Select "File | Open..." from menu.
  • If file being opened is not space delimited, the "Text Import Wizard" will appear to allow you to specify a delimiter.


    Figure 1: "Text Import Wizard" dialog box
    Return to Index


    2. Automatic file size detection

    You do not need to specify whether a file is large, file size will be detected automatically while opening the file. If row or column size exceeds spreadsheet capacity, Trend View will show a dialog box informing you the incremental sampling feature is active.


    Figure 2: Notification on activing sampling feature.
    Return to Index


    3. Automatic initial sampling

    The initial sampling interval is calculated dynamically from file size and spreadsheet capacity, such that the large dimension(s) of the file fits exactly in the spreadsheet. After it opens, you'll find file size and sampling interval information in the top most row (or left most column if columns were sampled.).




    Figure 3: Initial sampling results.
    Return to Index


    4. Custom re-sampling for detailed data

    Identify a region of interest in the data - you can try the various Trend View data visualization tools:

  • 2D line graph, bar graph, 2D/3D scatter plot, bubble plot
  • false colour analysis

  • Figure 5: Finding region of interest using 2D line graph

    Re-sample in this region

  • Select cells in region of interest (left mouse button)
  • Right-click, and select "Get detailed data" from menu
  • More data from the highlighted region will be loaded from file into spreadsheet (sampling interval is again calculated based on quantity of target records in file and spreadsheet capacity)

  • Figure 6: Specifying region to re-sample at higher resolution


    Figure 7: Results from re-sampling

    Return to Index


    5. Getting full resolution data (do not skip records from file)

    By selecting just one cell, and choosing "Get detailed data" from right-click menu, re-sampling will take place at full resolution, i.e. no rows (or columns, if file is wide) is skipped when loading from file.

    Note: The number of records loaded at full resolution is again confined to spreadsheet capacity, but this way you have access to a custom portion of the file instead of just the beginning.


    Figure 8: Re-sampling from one cell to get full resolution data from file.
    Return to Index


    File access with sampling is Read Only

  • Files opened via sampling cannot be saved, to prevent accidental modification of your data file. If you wish to preserve the spreadsheet as is, you can cut and paste its contents into a new spreadsheet and save that instead.
  • Return to Index