Benchmark
Benchmark
- k8s-gpu-scheduler-lab
A controlled comparison of Kubernetes GPU schedulers on the same workload traces, built on kwok so it runs without GPUs.
- slurm-rca-bench
A public incident-diagnosis benchmark for HPC schedulers — ten reproducible failure scenarios with measured ground truth and degenerate baselines.
- vLLM vs TensorRT-LLM: Inference Throughput
A measured throughput comparison of vLLM and TensorRT-LLM on Llama-3 70B, and why the faster engine is not automatically the right one.
- vLLM vs TensorRT-LLM: First Throughput Numbers on Llama-3 70B
Quick benchmark comparing vLLM and TensorRT-LLM throughput on a single A100 80GB with Llama-3 70B. Setup notes, raw numbers, and what I'd investigate next.