GPU Scheduling Landscape and mutex Semantics
Why GPU scheduling matters: how the default scheduler misses node, card, and topology placement, how HAMi v2.10 fills the gaps, and what mutex really means on kind.
HAMi
GPU sharing is moving from soft allocation to governance—scheduling decisions and runtime isolation can finally be reconciled.
HAMi Community Evolution: When AI Writes Code, What Makes an Open Source Community Valuable?
A long-term participant’s observation on HAMi’s growth (2021 open source, 2024 CNCF Sandbox, 2026 CNCF Incubating) and what makes an open source community valuable in the AI era — AI lowers the cost of producing code, but not the cost of building consensus. Tomorrow’s maintainers are consensus builders.
After HAMi's CNCF Incubating: From a Community of Code to a Network of Consensus
Code is cheap; consensus is the new scarce good.
Olares and HAMi: Desktop AI Workstation Inflection
HAMi moves from cluster to desktop with Olares.
From GPU utilization to productive GPU-hours.
From GPU to Token: The 8-Layer Observability Stack for AI Infrastructure
From GPU hardware, Kubernetes scheduling, inference engines to token cost — understanding the 8-layer observability architecture for modern AI infrastructure.
AI Infra Industry Trends: From Compute Bottlenecks to Ecosystem Evolution
A practitioner’s perspective on AI infrastructure trends: evolving bottlenecks, roles of CPU/GPU/scheduling, ecosystem shifts, and compute demand across training, inference, and Agent workloads.
Kubernetes as the GPU Control Plane for AI
Observations on the evolution of AI infrastructure control planes, focusing on HAMi v2.9, GPU scheduling, and Kubernetes resource models.
KubeCon EU 2026 Day One Observations
KubeCon Europe 2026 Day One: How Kubernetes is adapting to the AI infrastructure wave and the evolution of the GPU resource layer.