Exact Evaluation of Adaptive Prefix-Cache Continuations

In LLM serving with prefix caching, several request orders can incur the same immediate prefill work yet leave different cache states. We give an exact method to score a known, finite continuation from each state under least-recently-used eviction and entry-time charging of whole prefix paths. The continuation can select its next request using current hits. Equivalent starting states can share scores, and a sufficient persistence test can determine a score without replay. In one previously examined Mooncake tool/agent condition (128 blocks, decision output), the two shortcuts reduced full replays from 267,672 to 46,610. The complete-state and focused-state implementations used the same shortcuts and agreed on replay counts in each of eight conditions, as expected. The complete-state implementation used more mean native CPU in six of the eight. In the highlighted condition the focused-state implementation took 66.59 microseconds per decision versus 89.00 for the complete-state implementation. Local caller wall time fell 16.9%. The policy saved 0.204% of modeled prefill work at 128 blocks and 1.001% at 32 on Mooncake, and zero on two external episodes. We have not established an end-to-end benefit for an independent planner.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-12
DOI
https://doi.org/10.5281/zenodo.24177974
Primary Topic
Parallel Computing and Optimization Techniques
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Exact Evaluation of Adaptive Prefix-Cache Continuations

Mohammad Ali Seyed Mahmoud
Zenodo (CERN European Organization for Nuclear Research)
Parallel Computing and Optimization Techniques
preprint

Exact Evaluation of Adaptive Prefix-Cache Continuations

Mohammad Ali Seyed Mahmoud
preprint en

Abstract

In LLM serving with prefix caching, several request orders can incur the same immediate prefill work yet leave different cache states. We give an exact method to score a known, finite continuation from each state under least-recently-used eviction and entry-time charging of whole prefix paths. The continuation can select its next request using current hits. Equivalent starting states can share scores, and a sufficient persistence test can determine a score without replay. In one previously examined Mooncake tool/agent condition (128 blocks, decision output), the two shortcuts reduced full replays from 267,672 to 46,610. The complete-state and focused-state implementations used the same shortcuts and agreed on replay counts in each of eight conditions, as expected. The complete-state implementation used more mean native CPU in six of the eight. In the highlighted condition the focused-state implementation took 66.59 microseconds per decision versus 89.00 for the complete-state implementation. Local caller wall time fell 16.9%. The policy saved 0.204% of modeled prefill work at 128 blocks and 1.001% at 32 on Mooncake, and zero on two external episodes. We have not established an end-to-end benefit for an independent planner.

Zenodo (CERN European Organization for Nuclear Research)
Georgia Institute of Technology (US)
Parallel Computing and Optimization Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon