How are key collision situations handled in HashMap?

In HashMap, key collisions (collisions) are handled using chaining. This means that all elements that have the same hash code index in the bucket array are stored in a linked list or a tree (starting from Java 8) within that bucket.

When inserting, HashMap calculates the hash of the key and determines the bucket. If the bucket is empty, the element is simply added. If there are already elements in the bucket, it iterates through these elements to check if a key equal to the inserted one exists. If the key matches, the value is updated. If not, the new element is added to the chain.

Starting from Java 8, if the chain in a bucket becomes too long (more than 8 elements), the linked list is converted into a balanced tree (TreeMap), which improves search performance from O(n) to O(log n).