CCL-Bench packages traces and metadata to compute detailed compute, memory, and communication efficiency metrics, surfacing performance insights unavailable from end-to-end benchmarks.
Vidur: A large-scale simulation framework for llm inference.Proceedings of Machine Learning and Systems, 6:351–366
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
A flow-control framework for LLM inference derives necessary and sufficient stability conditions and experimentally improves throughput, latency, and KV cache stability over common baselines.
citing papers explorer
-
CCL-Bench 1.0: A Trace-Based Benchmark for LLM Infrastructure
CCL-Bench packages traces and metadata to compute detailed compute, memory, and communication efficiency metrics, surfacing performance insights unavailable from end-to-end benchmarks.
-
Flow-Controlled Scheduling for LLM Inference with Provable Stability Guarantees
A flow-control framework for LLM inference derives necessary and sufficient stability conditions and experimentally improves throughput, latency, and KV cache stability over common baselines.