# HBF: a read-oriented tier and its evidence boundary

Canonical: https://semiaifoundry.com/learn/atlas/hbf-read-tier-contract/
Revision: 2026-10-07
Rights: https://semiaifoundry.com/ai-atlas/licenses.html

Memory · Atlas v1.3.0

# HBF: a read-oriented tier and its evidence boundary

Distinguish an announced HBF interface/capacity specification from validated workload behavior.

[Explore in Atlas](https://semiaifoundry.com/ai-atlas/#hbf-read-tier-contract)Processor interface to NAND-backed capacity; logical topology, not a disclosed product floorplan

Learning objectiveDistinguish an announced HBF interface/capacity specification from validated workload behavior.

## The mechanism

High Bandwidth Flash proposes a NAND-based tier near the processor, alongside the existing memory hierarchy. SK hynix’s August 2026 announcement describes up to 512 GB and bandwidth grades of approximately 0.4–3.0 TB/s, with UCIe connectivity. These are reported specification boundaries, not proof of sustained throughput on a deployed inference system. A capacity and read-bandwidth headline leaves access latency, request granularity, concurrency, write service and host software to be examined. NAND-based capacity can be attractive for read-heavy state, but it does not acquire DRAM semantics merely by sharing a package or a fast link.

1.  Identify the source date and distinguish technical specification, sample, qualified component and shipping system.
2.  Map the processor link, HBF controller and NAND array as separate interfaces; use the same transfer direction and byte convention.
3.  Request latency, granularity, concurrency and read/write conditions before treating the headline rate as usable.
4.  Keep mutable working state and recovery requirements in the placement contract rather than assuming every tensor can move to flash.
5.  Measure application-level correctness, latency and energy on a named integration before claiming a deployment benefit.

## Science and assumptions

### Minimum bulk-read transfer time

$$
t_{\mathrm{transfer}}\geq\frac{V}{B}
$$

ttransfer ≥ V / B

Symbols and units

ttransfer

Bulk transfer time lower bound

s

V

Required payload volume

bytes

B

Assumed usable read bandwidth

bytes/s

Moving V payload bytes cannot take less than V/B at usable bandwidth B, even before initial access, scheduling or contention.

Assumptions Single direction with compatible decimal units; B is a sustained payload assumption, not automatically an advertised interface maximum.

## Engineering example

Synthetic payload-only calculation: reading 100 GB at an assumed sustained 1 TB/s requires at least 0.1 s. No setup, concurrency loss or other work is included. The assumed usable rate is not a measured HBF benchmark.

## Distinctions and tradeoffs

An interface/specification announcement and a measured shipping-system result are different evidence states.

More near-compute capacity may reduce remote movement, while page access, writes and integration requirements can move the performance bottleneck.

Large read-mostly parameter pools are a possible use case; they require a software and system plan that accounts for access behavior.

## Specifications and interfaces

Evidence snapshot

Announced · 4 August 2026

Reported specification: up to 512 GB; grades approximately 0.4–3.0 TB/s.

Interface

UCIe in the announcement

A common framework is not a drop-in compatibility or shipped-system certificate.

Still configuration-dependent

Latency, granularity, read/write behavior

The headline grades do not settle end-to-end service performance.

### Interfaces

Processor/HBF interface Input

Requests, completion, error and transfer semantics

NAND-backed capacity Output

Data with specified access and persistence behavior

## Check your understanding

### What does the announced HBF bandwidth range establish?

1.  A reported specification range that still needs workload validation
2.  Guaranteed HBM-equivalent random-access latency
3.  Proof that every GPU can attach HBF today

Answer and reasoning

A reported specification range that still needs workload validation

An interface/specification announcement and a measured shipping-system result are different evidence states.

## Evidence and scope

SK hynix and Sandisk identify the August 2026 milestone as specification release. The atlas has not verified a volume-shipping deployment from these announcements. Missing access metrics remain unknown rather than receiving SSD or HBM defaults.

Place read-only weights and mutable KV across specified tiers and inspect transfer paths; no HBF shipping-system benchmark or universal memory semantics are implied.

## Sources

-   [SK hynix — first HBF specifications at FMS 2026](https://news.skhynix.com/en/hbf-at-fms-2026/)
-   [Sandisk — release of first OCP HBF specification](https://investor.sandisk.com/node/8126/pdf)
-   [UCIe — introduction Q&A recap](https://www.uciexpress.org/post/introduction-to-ucie-webinar-q-a-recap)

[Published lesson data](https://semiaifoundry.com/ai-atlas/lesson-data/hbf-read-tier-contract-e7512c9ff78d70e4.json) · [Atlas releases and licensing](https://semiaifoundry.com/ai-atlas/licenses.html)

## Continue exploring

### Foundations

-   [Choosing SRAM, HBM, DDR, LPDDR or GDDR](https://semiaifoundry.com/learn/atlas/memory-interface-choice/)
-   [Die-to-die links and UCIe compatibility](https://semiaifoundry.com/learn/atlas/die-link-compatibility/)

### Related lessons

-   [HBF latency, transfer granularity and prefetch](https://semiaifoundry.com/learn/atlas/hbf-latency-prefetch/)
-   [Place the working set, not just the model](https://semiaifoundry.com/learn/atlas/memory-working-set-placement/)
-   [Qualify the memory system and retain useful work](https://semiaifoundry.com/learn/atlas/memory-system-qualification/)