Cloud Native Computing: A Computer Weekly Downtime Upload podcast

During the recent KubeCon/CloudNativeCon 2024 event in Paris, Computer Weekly met with Priyanka Sharma, executive director of the Cloud Native Computing Forum (CNCF), to record this podcast, looking into trends driving Kubernetes, inclusivity and the perceived gap between the cloud native community and senior IT management.

The tech industry is constantly pushing new technologies. Artificial Intelligence (AI) is the latest tech driving conversations across the tech sector. It is something those involved in advancing Kubernetes and cloud native architectures regard as an important technolology to support.

But like many tech innovations, there is a lot of industry hype associated wuth AI. Quoting Alan Greenspan, former chief of the Federal Reserve Board in the US, Sharma used her keynote to discuss “irrational exuberance” in technology innovation. Greenspan’s quote refers to the risk that some US stock prices are overvalued. In the tech sector and across society, Sharma points out that people get very excited about new technology innovation.

She says they invest in resources and spend time looking into concepts that are not 100% fully thought through. Some of these may not yield the results needed to give a return on this investment. “Identifying that something is irrationally exuberant is,” says Sharma: “When there's a bit of cognitive dissonance.”

Sharma is a self-taught software developer and she admits she has had ups and downs in her journey in software development. Her journey began when she was setting up her own startup. “We had to build our website. Someone had to get it going and that's how I learned.” Now, full time at CNCF (Cloud Native Computing Foundation), her main opportunities to drill down into software development happens in preparation for the twice-yearly KubeCon/CloudNativeCon events.

Kubernetes is being positioned as the technology to make AI workable. Container orchestration is being applied to manage graphics processing units (GPUs) for running machine learning and inference workloads. During her keynote presentation, Sharma gave a little demo showcasing the ease with which she could depoy a Large Language Model (LLM) application on her laptop using Kubernetes.

She says: “Over the last few years, I have actually really enjoy doing my keynotes. That's where I've made a conscious decision to immerse myself in a certain technology, learn as much as I can and develop a demo with the help of my team and then deliver it. That is often the way I get to stay engaged [in software development] while also running an organisation that has as large of a footprint as CNCF.”

In the video below, Enterprise Strategy Group analysts discusss the highlights of KubeCon 2024 in Paris.