One problem of performing clustering in high-dimensional data is that common distance metrics, such as Euclidean distance, do not perform as well as the number of dimensions becomes large. This is one of the issues caused by the Curse of Dimensionality, as the distance between any pair of points becomes less distinguishable as the dimensionality increases. One approach for dealing with high-dimensional data is to first reduce the dimensionality through a technique like PCA and then perform clustering on the principal components rather than the original data. Alternatively, algorithms such as DBSCAN are better suited to K-Means for identifying clusters in high-dimensional data.