AMU 2026 — Computer PYQ

To understand why option (b) is the defining feature of a clustered index, let's explore how it alters data storage on a physical disk:

• Physical Ordering (Statement b): A clustered index sorts and stores the actual data rows of the table on the disk based on the index key values. Because the physical data rows can only be sorted in one unique order across a disk volume, there can only be one clustered index per table.

Analysis of the Incorrect Options

Let us evaluate why the remaining options are incorrect or inaccurate definitions:

• Option (a) is Incorrect: While some database management systems technically permit a single NULL value in a unique index column, a clustered index typically acts as the primary key constraint (which enforces non-nullable entries explicitly via NOT NULL). Having multiple NULL entries defeats its primary lookup mechanism.

• Option (c) is Incorrect: A clustered index is remarkably fast for range-based queries (e.g., finding values between $10$ and $50$) because the data resides sequentially on disk. However, it is not always faster. For direct single-row exact match lookups or heavy insert/update heavy operations, a non-clustered index or a heap organization can perform better because shifting data physically on a disk introduces overhead.

• Option (d) is Incorrect: A clustered index can absolutely be created on multiple columns (known as a composite clustered index). There is no strict rule stating that these individual columns must be inherently unique; the database engine can append an internal unique identifier (like a "uniqueifier" in SQL Server) to keep non-unique rows organized.

Summary Checklist Comparison