Operations | Monitoring | ITSM | DevOps | Cloud

The default Kubernetes scheduler was built for stateless services: web servers, APIs, databases. It schedules a pod, checks that a node has enough of whatever resources were requested, and binds it. For CPU and memory, that model works fine. For GPUs, it falls apart in three specific ways. First, GPUs are treated as an opaque integer resource.

Every vendor in networking has an agent story right now. The useful question for an operations leader is which of those agents can plan, act, and verify against a trustworthy model of the network, and which are assistants that retrieve and suggest, then leave the decision to a person. The direction of travel is settled.

Enterprise messaging is not a solved problem sitting still. The last five years have reshaped the technology landscape in ways that are still working their way through enterprise architecture decisions: Kafka's dominance in event streaming, the rise of cloud-native managed messaging (Amazon MQ, Azure Service Bus, Confluent Cloud), the democratization of the Kafka protocol across competing implementations, and now the early emergence of agentic AI as a new category of messaging consumer.

Discover how Kentik and Infoblox combine network observability, authoritative network data and AI-driven operations to simplify hybrid multi-cloud infrastructure.