Lesson Learned from the Building of a Terascale AI Server Infrastructure
summary
Describes building Kasetsart University's ARES cluster (4-node servers with NVIDIA V100 GPUs on a 100Gbps fabric) and the Poseidon DGX-1 (8× V100s, ~1 petaflop) to support deep learning and HPC research. Covers hardware and software co-design decisions, implementation lessons, challenges encountered and solutions used to provide an AI service infrastructure for advanced researchers.
AI-written summary · about
chapters (11)
- 0:00Start
- 0:45Vehicle damage prediction
- 1:10Plant counting
- 1:39Face recognition & Emotion detection
- 2:06Food classification
- 4:16Processing Steps
- 6:20Design consideration
- 7:52Computing Capability
- 9:17Software Installation
- 9:59Mixed Workload
- 13:09Benchmarks
related talks
28mScaling AI Infrastructure
28mScaling agentic AI: From infrastructure to enterprise outcomes
22mResearch SuperCluster(RSC): Accelerating AI Research at Scale
21mDesigning Scalable Networks for Large AI Clusters: Challenges and Key Insights
01h 15mTutorial: Build, Operate, and Use a Multi-Tenant AI Cluster Based Entirely on Open Source
41mBuilding & Operating a Large-scale HPC AI Cluster on Kubernetes
25mEmpowering AI-Driven Drug Discovery: Overcoming Challenges in Building a ML Platform on Kubernetes
16mUsing AI and Computer Technologies to Enhance IT Infrastructure