LLM serving should triage by five-resource analytical floors and wall ordering, not grid search; on 16×H20, TP16 is capacity-capped at ~70 while EP+DP attention reaches ~644 concurrent 8K requests.
RFC: Data parallel attention and expert parallel moes.https://github.com/ vllm-project/vllm/issues/16037, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.PF 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Think Before You Grid-Search: Floor-First Triage for LLM Serving
LLM serving should triage by five-resource analytical floors and wall ordering, not grid search; on 16×H20, TP16 is capacity-capped at ~70 while EP+DP attention reaches ~644 concurrent 8K requests.