Announcing MCP Toolbox Java SDK v1.0: Agentic data access for the enterprise
cloud.google.com·6d ago
Cloud & Infrastructure
Google Cloud Blog·Aug 5·5 min read
Scaling agentic AI: How UiPath built its high-performance GPU platform on AI Hypercomputer
TL;DR
UiPath has developed a high-performance GPU platform on Google Cloud to support its transition to agentic AI, enabling autonomous agents to make decisions and execute complex business processes. This infrastructure allows for efficient orchestration of GPU resources, balancing training and inference workloads effectively.
✦ Why It Matters
Consider adopting a similar GPU orchestration strategy to enhance your AI infrastructure today.
Key Takeaways
- Implement a shared GPU resource model to manage workloads efficiently.
- Use Dynamic Workload Scheduler to plan compute capacity in advance.
- Utilize A3 VM instances for training and G4 VM instances for inference.
- Focus on reducing operational overhead by centralizing GPU management.
How It Works
UiPath's shared GPU fleet is managed by its machine learning services platform, which prioritizes workloads across teams and time windows. This allows for real-time inference during busy hours and batch training during off-peak times, maximizing GPU utilization and reducing idle capacity.