Draw the internal layout of an 8 KB PostgreSQL Slotted Page (`PageHeader`, `ItemIdData` line pointers, free space, `HeapTupleHeader`, null bitmap, aligned.

A strong answer

A strong candidate frames Draw the internal layout of an 8 KB PostgreSQL Slotted Page (PageHeader, ItemIdData line pointers, free space, HeapTupleHeader, null bitmap, aligned. around the exact memory layout, compiler semantics, and hardware execution model covered in [8 KB Slotted Pages, Line Pointer Indirection (ctid), Column Alignment Padding, and TOAST]. Explain why line pointers grow down while tuples grow up, how HeapTupleHeader costs 23 bytes per row, how NULL columns occupy 0 data bytes via the null bitmap, and why SELECT * on tables with TOASTed columns hurts I/O.

You store 500 million rows in a PostgreSQL table: CREATE TABLE telemetry (flag1 BOOLEAN, id BIGINT, flag2 BOOLEAN, ts TIMESTAMPTZ, flag3 BOOLEAN, val DOUBLE PRECISION). You calculate the raw data size as 1 + 8 + 1 + 8 + 1 + 8 = 27 bytes/row, yet each row actually consumes 72 bytes on disk (wasting 10+ GB of RAM and SSD!), and when you run UPDATE telemetry SET val = 9.9 WHERE id = 42, you wonder: how can the database slide tuples around inside an 8 KB page to defragment free space without having to update 6 different B+ Tree indexes that point to row 42?

When engineers encounter this bottleneck in production, a surface-level API description rarely explains what went wrong. In real workloads, every line of code executes against concrete hardware constraints: fixed-width memory cells, CPU cache lines, kernel descriptor tables, network socket buffers, or accelerator memory buses. Without a clear mental model of 8 KB Slotted Pages, ItemId Line Pointer Indirection ((page_id, slot_id)), Column Padding, and TOAST, debugging degrades into guesswork—tuning timeouts, adding retries, or scaling instances blindly while the underlying architectural flaw remains untouched.

In a production interview, always connect the specification rule to its concrete runtime consequence: state what lives on the stack versus the heap or static storage segment, how the compiler or operating system validates alignments, lifetimes, and error codes, and what happens under adversarial or boundary inputs (such as zero-length buffers, integer overflow, concurrent mutation, or partial I/O). Backing your explanation with a small trace of register/memory state transitions immediately separates senior systems engineers from candidates who only memorized syntax.

What the interviewer is listening for

• Mastery of core invariant: Trace the binary layout of an 8 KB Slotted Page (pd_lower line pointers growing down, pd_upper tuples growing up) and ctid indirection.
• Mastery of core invariant: Eliminate alignment padding waste ('Column Tetris') by ordering table columns from 8-byte down to 1-byte and variable-length types.
• Mastery of core invariant: Explain how TOAST compresses and slices values > 2 KB and why SELECT * triggers unnecessary TOAST I/O.
• Hardware and ABI awareness: Explaining how sizes, alignments, calling conventions, or kernel boundaries behave on real 64-bit targets rather than relying on implementation-defined assumptions.
• Defensive engineering: Naming the exact compiler flags (-Wall -Wextra -Werror), static analyzers, or runtime sanitizers (-fsanitize=address,undefined,thread) that verify correctness in CI.

Common mistakes

• Explaining only surface syntax without tracing the underlying memory addresses, ownership lifetime, or kernel/user-space state transitions.
• Assuming undefined behavior (such as signed overflow, uninitialized reads, strict-aliasing violations, or dangling pointers) will "just work" because it passed an unoptimized -O0 debug build.
• Ignoring return codes, partial reads/writes, alignment padding, or cleanup paths when an allocation or syscall fails midway through a function.