amp-buffer-caching · research · also Adaptive Multi-Policy caching, Program-context-specific buffer caching

AMP buffer caching

AMP is a research buffer-caching framework that selects replacement policies and adapts cache partitions using the program contexts issuing I/O.

Related Observatory article: Read on hesela.com

Requests are grouped by call-stack context, rather than only by file or process. Detected looping contexts use MRU partitions; the design discusses ARC as the default and adapts partition sizes using randomized competition. Repeated accesses alone do not distinguish looping from temporal clustering. [1]

The reference-recency score uses the previous occurrence's position p in the oldest-to-newest list of distinct previously seen blocks: p/(n-1). First occurrences are excluded; a repeated block with n=1 receives 0.5. The reported loop threshold is 0.4, not a universal workload boundary. [2]

Hesela's synthetic examples [1,2,3,1,2,3] and [1,1,2,2,3,3] have identical lengths, per-block counts and three repeat observations. Their mean scores are 0 and 5/6 respectively. With no repeat observations the mean is undefined, not zero. These calculations distinguish order from frequency; they are not cache hit rates or measured speedups. [3]

The Linux 2.6.8.1 implementation retains a CLOCK-like default partition, so the design's ARC discussion must not be silently substituted for the kernel baseline. The evaluation's PCC* comparator combines PCC detection with AMP cache management, rather than reproducing the complete PCC system. [2]

Relations

related
Adaptive Replacement Cache, Least recently used replacement, Locality of reference, Program-counter-based classification (PCC)

Limitations

  • The educational calculation omits context collection, sampling, partition competition and actual I/O; it does not replicate AMP's kernel prototype. [3, 2]

Citations

  1. Zhou, von Behren and Brewer. AMP: Program Context Specific Buffer Caching. USENIX ATC 2005, sections 1-3. (accessed 2026-11-21)
  2. Zhou, von Behren and Brewer. Program Context Specific Buffer Caching with AMP. UCB/CSD-05-1379, April 2005, sections 3.1, 3.3 and 4.2. (accessed 2026-11-21)
  3. Hesela. Bounded reference-recency calculation for synthetic integer block traces. (accessed 2026-11-21)

Further reading

  1. Hesela. Reference-recency boundary cases and independent last-occurrence oracle. (accessed 2026-11-21)
Sources & caveats
  • Primary-source synthesis and a transparent arithmetic illustration, not a modern storage benchmark or independent replication of reported performance.
  • Historical implementation and comparison qualifications are taken from the full technical report, not inferred from the abstract. [2]

JSON: /data/terms/amp-buffer-caching.json

Published by Hesela team · Updated

Report a correction · Editorial standards · Reuse guidance

Cite this page
Hesela team. AMP buffer caching. Hesela. 2026-11-21. https://hesela.dev/terms/amp-buffer-caching/