What characteristic of a Non-Unique Primary Index (NUPI) can lead to performance issues?

A Non-Unique Primary Index (NUPI) allows for duplicate values in indexed columns, which can lead to skewed data distribution across the system's processing nodes. In Teradata, the data associated with NUPI values is potentially spread unevenly across the disks, especially when some values are more common than others. This uneven distribution can cause certain nodes to become overloaded with data and queries, while others may be underutilized.

When data is skewed, it can lead to performance degradation due to increased processing time on the nodes with higher data volumes, as they take longer to execute queries. As a result, query response times can increase, and resource utilization can be inefficient. Addressing this skew typically requires careful data modeling and indexing strategies to ensure more balanced data distribution.