#006 - Kubernetes for Humans Podcast with Nick Jones (EscherCloudAI)
Sep 27, 2023
auto_awesome
Explore Nick Jones's journey from 3D graphics to cloud computing, the evolution of container orchestration technologies with a focus on Kubernetes pitfalls and organizational challenges, AI and Kubernetes trends including GPU utilization and job scheduling tools, and the community-driven evolution of the 'Sausage Cloud' project from a nuclear bunker to a public cloud deployment.
Cultural shift crucial for successful Kubernetes adoption, emphasizing cohesive engineering teams and alignment between developers and infrastructure.
Future of Kubernetes involves increased consumability, seamless integration with diverse workloads, and emphasis on practical application over latest API specifications.
Deep dives
Nick Jones's Background and Journey to Kubernetes
Nick Jones, head of cloud native and platform engineering at Azure Cloud AI, shared his journey into computer science, starting with a background in 3D graphics during university. His early work in a traditional sys admin role led to a passion for getting computers to communicate, with a shift towards cloud computing. With experiences at Sun Microsystems and a UK startup, Nick delved into community involvement, emphasizing the role of cultural shifts and teamwork in successful Kubernetes adoption.
Challenges in Implementing Kubernetes in Companies
Nick highlights the significant cultural shift required when transitioning to Kubernetes, stressing the importance of cohesive engineering teams that collaborate effectively. Traditional siloed organizational structures often hinder Kubernetes adoption, emphasizing the need for alignment between developers and infrastructure teams. Nick advises starting with understanding the problems to be solved and gradually introducing Kubernetes in a way that aligns with specific organizational needs.
AI and Kubernetes Integration for GPU Utilization
At Azure Cloud AI, Nick works on providing a Kubernetes deployment optimized for GPU-intensive workloads, particularly in the realm of AI. He notes the overlap between traditional HPC and AI, highlighting the evolving landscape of job scheduling in Kubernetes for AI tasks. Projects like Q and Volcano aim to enhance Kubernetes scheduling capabilities, with Nick foreseeing continued advancements in infrastructure support for AI workloads.
Future Trends in Kubernetes and Infrastructure Evolution
Looking ahead, Nick envisions a future where Kubernetes becomes more consumable and integrates seamlessly with various workloads. He anticipates a shift towards building easier-to-use platforms on top of Kubernetes, focusing on practical application rather than the latest API specifications. Nick predicts Kubernetes becoming a foundational infrastructure component, with an emphasis on tooling that aids adoption of diverse workloads.
Nick is a CNCF and OpenUK Ambassador, currently employed as Head of Cloud Native and Platform Engineering at EscherCloudAI, where he's helping to build a next-generation managed Kubernetes service tailored for AI workloads backed by a sustainable approach to running infrastructure. He is also a serial meetup organizer, including Cloud Native and Kubernetes Edinburgh, CNK Manchester, as well as larger events such as Kubernetes Community Days UK. Nick is passionate about new technology, but simultaneously romantic about the old with a penchant for decrepit Silicon Graphics and Sun Microsystems hardware.
-------------------