> [!NOTE] Tid Scan > <table> > <tr> > <td width="25%"><img src="assets/ex_tidscan.png"></td> > <td>The fastest possible scan. It fetches a row directly by its physical location (Block number and Offset). This is used when the query explicitly specifies the <code>ctid</code> system column, bypassing indexes and sequential scans entirely.</td> > </tr> > </table> > > ```sql > -- Fetching a tuple directly by its block and offset > EXPLAIN (ANALYZE, COSTS, BUFFERS, VERBOSE) > SELECT * FROM animals WHERE ctid = '(0,1)'; > ``` > > ```text > Tid Scan on public.animals (cost=1.00..5.01 rows=1 width=27) (actual time=1.004..1.004 rows=1 loops=1) > Output: id, name, species_id, created_at > TID Cond: (animals.ctid = '(0,1)'::tid) > Buffers: shared hit=1 > ``` > > <table> > <tr> > <td rowspan="5" width="25%"><img src="assets/ex_tid_scan.svg"></td> > <td><b>Performance</b></td><td>The highest performance possible for a single-row lookup; bypasses the index entirely to perform a single-page read.</td> > </tr> > <tr><td><b>Factors</b></td><td>Number of tuple IDs and whether the physical location is known in advance.</td></tr> > <tr><td><b>Cost</b></td><td><code>cpu_tuple_cost * number of tuple IDs</code></td></tr> > <tr><td><b>Operates on</b></td><td><a href="Structures/Tuple">Tuple</a></td></tr> > <tr><td><b>Workloads</b></td><td><a href="Workloads/IO/DataFile/DataFileRead">IO: DataFileRead</a>, <a href="Workloads/LWLock/Buffers/BufferContent">LWLock: BufferContent</a></td></tr> > <tr><td colspan="3"><b>Description</b>: Scans a table using tuple IDs.</td></tr> > </table>