Assume the following:

You are given a data set with 100 records and are asked to cluster the data. You use K-Means to cluster the data, but for all values of K, 1 K 100, the K-Means algorithm returns only one non-empty cluster. Then, you apply incremental versions of k-means but obtain exactly the same result.

How is this possible? Explain your answer.