Take the 2-minute tour ×
Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It's 100% free, no registration required.

I am just wondering - is there a name for these kinds of graphs? The associated text that comes with the graph says the following:

enter image description here

Figure 2 presents a view of the distribution of correctable errors over DIMMs. It plots the fraction of errors made up by the top x percent of DIMMs with errors. For all platforms, the top 20% of DIMMs with errors make up over 94% of all observed errors. For Platform C and D, the distribution is even more skewed, with the top 20% of DIMMs comprising more than 99.6% of all errors. Note that the graph in Figure 2 is plotted on a log-log scale and that the lines for all platforms appear almost straight indicating a power-law distribution.

I don't understand how this figure was plotted nor the description that comes with it because none of the values are visible on a log-log scale. It does not look like a conventional CDF curve to me. I have some data as follows:

Platform | Device | #Failures
Type 1   | A      | 1123
Type 1   | B      | 362
Type 1   | C      | 123
Type 1   | D      | 21

Now to get a CDF, I would take the #Failures column and plot its CDF in which I will get a graph with:

  • X-Axis: #Failures
  • Y-Axis: Cumulative Distribution of devices

So I figured I will normalize the #Failures column, draw the CDF of the normalized values Then I will get a graph with:

  • X-Axis: % of Failures
  • Y-Axis: Cumulative Distribution of devices

And then flip it? But I still don't understand what this new graph is telling me that the conventional CDF is not showing. Can someone shed some light onto this?

share|improve this question
    
This looks like an empirical CDF of the fraction of dimms with correctable errors with both the x and y axes in log-scale. –  EngrStudent Mar 24 at 3:21

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.