Data analysis is a process of inspecting, cleansing, transforming, and modeling data with the goal of discovering useful information, informing conclusions, and supporting decision-making.

Test Case Study:

This case study is an effort to study the effect on an Application when it is exposed with (i) No defender is running, (ii) Minimum Policy with CNNF, and (iii) Hard End Policy with WAAS. In this study we are using the TCP delta time, based on the TCP/IP three-way handshake request / response timings. The study is using the Data Analytics methods to analyze the results obtained.

Container Network

In this test case we wanted to test whether TCP/IP 3-way handshake when we are running the three different case scenarios as given above: We did discard any TCP, (i) retransmission, (ii) spurious retransmission, (iii) lost segment, (iv) ack lost segment between the request and response. We collected only the (v) Non-zero TCP time Delta of those interactions.

A. Analysis of Data by way of Visual Inspection

We first plotted the data and check if we can get some clues on the data collected.

TCP Time Delta plots

TCP Delta Time Plots for Idle case

TCP Delta Time Plots for 'With Defender' case

TCP Delta Time Plots for 'With No Defender' case

B. Analysis By Comparing the Data Sets Means using K-Means Clustering

if we cluster the data sets into 2 clusters we should get the K-mean and that can be used to find the distance between the means to tell how much is the worst time difference in seconds for each case scenario we are studying. As K-mean clusters circle around the majority of the data, thus showing the worst case scenario of the centers.

K-Means Clustering between Idle and No Defender

K-Means Clustering between Idle and With Defender

K-Means Clustering between With Defender and Without Defender