educative.io

Educative

Trade-off for the small timeout value illustration is not understandable

Illustration for Trade-off for the small timeout value is not understandable,
please explain the simulation what is the false positives?
How does it depict Trade-off for the small timeout value?


Course: https://www.educative.io/collection/10370001/4891237377638400
Lesson: https://www.educative.io/collection/page/10370001/4891237377638400/5770111284150272

Hello Afaque,
A false positive means the node was not dead, but it is perceived to be dead (falsely perceived) due to the expiration of the small timeout value. Please follow the following confusion matrix notation to understand it.
Confusion matrix
Let’s take the following assumption, the Alive node is a negative class, and the dead node is a positive class.

Alive node → negative class
Dead node → positive class

Here is the confusion matrix,

False negative (FP): If a dead node (actual/reality/ground truth) is perceived as alive (negative), we call it a false negative because we perceived false of what it actually was.

True negative (TN): If an alive node is perceived as alive (negative), we will call it a true negative because we perceived what it actually was, we truly perceived.

False positive: If an alive node is perceived as dead (positive), we call it a false positive because we perceived false of what it actually was.

True positive: If a dead node (actual/reality/ground truth) is perceived as dead (positive), we call it a true positive because we perceived what it actually was, truly perceived.

The small timeout value benefits in a way that we don’t have to wait long for crashed nodes that wouldn’t respond no matter how long we wait for them. While the small timeout value has a drawback in that the nodes that were not dead but were slower in responding in that short time span are considered dead.

Hope the above explanation helps. Please ask if there is any other question. Thanks!

1 Like