ShiftWing · LocalForge · Independent research

Dingal AI Research Frontier AI on consumer-grade hardware
Menu

Treating VRAM, RAM and NVMe as one memory hierarchy

Experiments with direct storage, learned expert locality and predictive prefetch turned three separate resources into one managed path.

The storage experiments asked whether routed experts could move through graphics memory, system memory and fast storage without making correctness depend on where a weight happened to live.

Method

  • Build a batched direct-storage path with an explicit cache.
  • Measure route locality at equivalent cache capacity.
  • Add prediction only behind a safety mechanism that could fall back to demand loading.
  • Compare every treatment with a control that used the same inputs and output checks.

Findings

The routes showed enough locality for a learned cache to be useful. Prefetch could overlap some waiting when its predictions were timely, but prediction was not reliable enough to become a correctness dependency.

Decision

These experiments validated the mechanism, not a universal performance claim. The result depends on workload locality and the relative behaviour of storage, memory and compute.


Cite this page: Dingal AI Research. “Treating VRAM, RAM and NVMe as one memory hierarchy”. 25 July 2026.

Related

Report

Pushing experts below four bits

Low-bit expert sidecars reduced storage pressure, but the quality study made clear that this is a trade to qualify rather than a free saving.

Dingal AI Research