# Table Operations ![[assets/ex_seqscan.png|256]] In the Directed Acyclic Graph (DAG) of a query plan, **Table Operations** are the **Leaf Nodes**. They are the primary ingestion points where the engine interacts directly with the physical storage of a relation. ### The Architectural Role Every query must begin by pulling data from a base source. Table operations are responsible for: 1. **Opening the Relation**: Identifying the physical file on disk. 2. **Page Ingestion**: Coordinating with the [[Manuscript/06 - Resource Management & Processes/6.2 - Shared Buffers (The Warming Rack)|Buffer Manager]] to bring 8KB pages into memory. 3. **Tuple Extraction**: Iterating through the [[Manuscript/02 - Physical Storage & MVCC/2.2 - Tuple (The Physical Suitcase)|Tuples]] within those pages and checking their visibility against the current transaction snapshot. ### In the Explain Plan Table operations typically appear at the deepest indentation of an `EXPLAIN` output. They represent the "raw material" flowing into the rest of the execution pipeline. ```text -> Seq Scan on animals (cost=1.00..34.50 rows=2450 width=18) ``` --- ![[Operations/Table/SeqScan]] --- ![[Operations/Table/TableFuncScan]]