Observability in Federated & Distributed Environments
Andreas Spanner, Jeyaramachandran Paulraj
As environments, workloads and AI training becomes more and more distributed so does the need for tracing, logging and monitoring to support safety, security, availability and resilience. This session introduces a Validated Pattern approach to Edge Observability where CTO's, product teams, engineering, system integrators, ISVs and customers come together to ease deployment of a common observability infrastructure. Apart from infrastructure monitoring with regards to compute and memory utilization a leading practice also supports metrics around power consumption to support sustainabilty efforts, as well as the enabling of workload specific observabilty. This session is also an open invitation for interested parties to join to evolve our open source approach to making observability accessible to everyone interested in edge, distributed and federated environments. We will be walking through the associated git repo, the components and a roadmap of planned features.
Cloud, Edge, and Platform Technologies
Tikona Room (Chanakya Building / School of Business)