Databases Reference
In-Depth Information
Figure 3.7
Scatter Plot of ResponseTime.csv
simulations. These omissions were not detected until results were viewed in a
similar scatter plot.
Change the Y axis to AvgPgRsp, the Z axis to FileSize, and the category
to Server.
Numerous outliers are visible among the observations at the 250 TPS level.
After review, these turned out to be simulations run without first returning the
servers to a predetermined initial state - again a mistake made by the technician.
They obviously need to be removed from the dataset before continuing the
analysis. The higher page response rates at the upper TPS and FileSize settings
were valid measurements reflecting degradation of the server at these levels. The
phenomenon observed in the plot is frequently referred to as the “hockey stick”.
Exercise 3.6
Using the ResponseTime.csv dataset in VisMiner:
a. Use the parallel plot to extract a subset named “outliers” of the invalid
observations.
b. In the Control Center, create a subset of valid observations using the
difference between the full dataset and the outlier set.
c. Name the dataset “ValidResponseTime.csv”.
