Case studies
Real environments.
Real technology outcomes.
From autonomous driving and finance to enterprise AI, explore how teams put heterogeneous compute sharing and scheduling into practice.
How SeresCloud scaled AI inference to 10 billion daily tokens with Kubernetes and HAMi
10B+Daily tokens on SeresCloud
Read the case studyHow Zhuoyu Technology raised GPU allocation above 95% on Kubernetes with Koordinator
>95%GPU allocation rate
Read the case studyHow Sangfor built an enterprise AI compute platform with Kubernetes, Volcano and HAMi
>3×Improvement in GPU utilization
Read the case study
CETC Cloud uses HAMi to build shared heterogeneous GPU infrastructure for a portable knowledge base
15×Development of environmental loads
Read the case studyOne heterogeneous compute pool, two workloads: China Merchants Bank unifies training and inference on Kubernetes
60%+Average accelerator utilization
Read the case studyChina Merchants Bank improves AI compute scheduling on Kubernetes with HAMi
100%Hardware resource pool utilization
Read the case study
GenAI for 200 million users: how SNOW orchestrates 1,000 GPUs to handle 700% traffic spikes
91%Reduction in average failure recovery time
Read the case studyNIO improves GPU utilization for autonomous driving workloads with HAMi
30%Reduced while imitating load GPU
Read the case study
Scaling machine learning infrastructure with Kubernetes and HAMi GPU virtualization
≈3×Increased utilization of the platform GPU
Read the case study
DaoCloud builds a flexible GPU cloud with HAMi
57%Average GPU utilization
Read the case studyEffectiveGPU: SF Technology’s heterogeneous AI virtualization and pooling solution built on HAMi
90%GPUs saved in production model serving
Read the case studyPREP EDU enables efficient GPU orchestration for AI inference
>80%GPU infrastructure optimized by HAMi
Read the case study