The New Stack Podcast Do All Your AI Workloads Actually Require Expensive GPUs?
Dec 18, 2025
Joining the conversation are Andrei Gueletii, a Google Cloud Technical Solutions Consultant; Pranay Bakre, a Principal Solutions Engineer at Arm; and Gari Singh, a Product Manager for Google Kubernetes Engine. They dive into why Google's Axion CPUs offer a compelling alternative to traditional GPUs, emphasizing cost-efficiency and energy savings. The guests discuss the flexibility of custom machine shapes in GKE, the shift to ARM architecture, and how many AI workloads can run more efficiently on CPUs. Insights into deployment at the edge and optimizing resource allocation add to the appeal.
AI Snips
Chapters
Transcript
Episode notes
Axion Targets Cost And Power Efficiency
- Google built Axion CPUs to deliver higher performance at lower cost and power for large-scale cloud workloads.
- Customers can get more compute and energy efficiency by shifting certain workloads from x86 to ARM-based instances.
Right-Size With Custom Machine Shapes
- Use N4A custom machine shapes to right-size memory and CPU per workload instead of fixed sizes.
- Tailor instance shapes to microservices and general-purpose workloads to improve price performance.
Platform Engineering Meets FinOps
- Platform engineering teams must optimize centralized platforms for price performance and FinOps as they scale.
- Choosing cost-effective compute under the platform improves developer experience and overall efficiency.