Predicting LLM Inference Latency: A Roofline-Driven ML MethodSaki ImaiRina Nakazawaet al.2024NeurIPS 2024
Best-Effort Power Model Serving for Energy Quantification of Cloud InstancesSunyanan ChoochotkaewTatsuhiro Chibaet al.2024MASCOTS 2024
Advancing Cloud Sustainability: A Versatile Framework for Container Power Model TrainingSunyanan ChoochotkaewChen Wanget al.2023MASCOTS 2023
Sustainable Computing: Measuring Application Energy Consumption in Kubernetes Environments with KeplerMarcelo AmaralSunyanan Choochotkaew2024KubeCon EU 2024
Make Kubernetes Networking Ready for world class AI and HPC workloadsSunyanan ChoochotkaewGaurav Singh2022Kubecon + CloudNativeCon NA 2022