Is this the same as a sorted file?

No. Direct slot mapping and ordering existing records by key are related but distinct organization choices.

Does it eliminate all indexes?

No. A system can still need indexes for other search fields or a mapping from external identifiers.

Are empty slots necessarily errors?

No. They can represent valid positions whose records are currently absent.

AI Glossary / Data engineering and architecture / Intermediate

Array-sequenced organization

Definition

Array-sequenced organization stores fixed-length records at positions determined by numeric array indexes. In Atzeni and colleagues' database teaching usage, adjacent storage blocks contain slots, and an index identifies a record's position. A business key can be mapped to that index, but the two need not coincide.

What it means

A simple design might reserve one slot for each equipment number in a bounded range. Record 105 can then be placed in a predictable slot instead of being found by comparing its key with many other records.

This is a physical organization concept, not a requirement that every database table be implemented as a programming-language array. It also differs from merely sorting stored records by a key. A sorted file may contain only existing records; a direct array arrangement can reserve positions for missing key values too.

The term is less common in modern product documentation than terms such as relative file or direct addressing. Those provide related concrete mechanisms, but their detailed formats and guarantees should not be assumed identical to every textbook description.

How it works

Choose an index range and a rule mapping each valid key to a slot. In a simple integer-key model, key k maps to slot k minus the lowest valid key. With equal-sized slots, the slot's offset can be calculated from its index and width.

The storage layer must distinguish occupied slots from empty ones. Finding a slot does not establish that a live record exists there. Insertions, deletions and updates therefore need an occupancy convention and appropriate coordination.

Fixed slot width makes offsets predictable, but useful record contents need not always fill the entire slot. IBM's relative-file documentation, for example, describes fixed-length slots accommodating records up to a specified maximum size.

On secondary storage, locating an offset still involves pages, caching and input/output. Constant-time index arithmetic is not a promise that every disk access takes the same elapsed time. Sparse key spaces, growth and oversized records can make another organization more suitable.

Three practical examples

A bounded device registry uses equipment numbers mapped to fixed slots when the number range and record format are stable.
A simulation stores one state record per indexed cell in a fixed grid, making the position mapping explicit.
A legacy relative-file application retrieves a record by its slot number and checks that the slot contains an active record before using it.

Think of it this way

Think of a wall of numbered pigeonholes, with one position for every allowed number. Knowing a number tells you where to look, but some pigeonholes may be empty.

The analogy highlights the space tradeoff: a million possible numbers require a different design judgment from a few hundred densely used numbers. Real storage also has page boundaries, metadata and recovery mechanisms. The picture explains addressing, not overall reliability.

Worked example

Calculating a position and checking occupancy

Consider an invented file for keys 100 through 199 inclusive. It reserves 100 slots, each 64 bytes wide, with a 128-byte header. Use zero-based slot indexes and assume no additional alignment or page overhead in this toy calculation.

Key 137 maps to index 137 minus 100, or 37. Its slot begins at byte offset 128 plus 37 times 64, which is 2,496. The reserved slot area occupies 6,400 bytes, so the file layout described here totals 6,528 bytes including the header.

If only 20 keys have live records, only 20 of the 100 slots are occupied. Locating key 137's slot still requires an occupancy check: its calculated position exists even if the record has been deleted or was never inserted.

Key 205 lies outside the defined range. Applying the arithmetic without bounds checking would not make it a valid record address. The implementation needs a stated growth or rejection policy, not just a formula.

Illustrative case study

A fictional facilities team uses a compact equipment-number range for a position-based registry. Later, imported identifiers contain large gaps and numbers far beyond the original range. Expanding the direct layout to cover every possible number would reserve mostly unused slots.

The team compares maintaining a separate dense internal index with adopting a keyed storage structure. It evaluates the extra mapping lookup against the wasted space and changes to the import process. It also checks that deleted slots cannot be mistaken for current equipment.

Layout suitability depends on identifier distribution. A design that works well for a bounded dense range may become unsuitable when an organization changes how it assigns keys.

Common misconceptions and mistakes

Array-sequenced means records must be scanned from the beginning. A principal attraction is calculating a known slot directly.
Every integer identifier is suitable for direct array storage. A huge sparse range can waste space or exceed practical bounds.
A calculated address proves a record exists. Occupancy, validity and bounds must still be checked.

Business use

This concept helps explain storage tradeoffs in legacy systems, specialized registries and performance-sensitive data structures. It matches access patterns and identifier ranges to predictable storage.

Before choosing such an arrangement, estimate the possible key range, expected occupancy and maximum record size. Include expansion and deletion behavior. Compare the cost of direct slots with an index that stores only present keys, using the real workload rather than arithmetic alone.

For AI and analytics pipelines, separate storage positions from lasting business identifiers. A reused slot may refer to a different entity later. Preserve stable identity and version information when exporting records, otherwise an efficient local lookup can become a misleading link across datasets.

Check your understanding

With keys starting at 500 and zero-based indexing, which slot corresponds to key 507, and what must be checked before reading its contents as a record?

Answer: Slot 7. Check the key bounds and whether that slot contains a valid occupied record.

Related terms

Access Method
Physical storage
direct addressing
array index
relative file
record slot
sequential organisation
sparse data

Learn next

Read direct addressing and relative files. Sketch a ten-slot layout containing three records, then describe lookup, deletion and an out-of-range insertion without assuming that every slot is occupied.

Sources

Atzeni, Ceri, Paraboschi and Torlone, Database Systems, Chapter 9 author slides, pages 75 and 77: https://dbbook.inf.uniroma3.it/BD09eng.pdf. Directly consulted primary treatment of array sequential structures, fixed-length tuples, indexed access and deletion slots.

IBM, Creating relative files: https://www.ibm.com/docs/en/txseries/9.1?topic=files-creating-relative. Related product implementation, not an assertion that every relative-file format equals the textbook organisation. The byte-offset calculation and facilities case are original illustrations.

Previous: Array Next: Article (part-of-speech)