By Guest: Michael Maxey (LinkedIn)
Company: Zededa (Twitter)
Show: Let’s Talk
While there are a lot of managed Kubernetes services nowadays, one company is differentiating itself by focusing on offering a small-footprint Kubernetes-managed service that can be deployed on the edge. Zededa has partnered with SUSE using EVE-OS and K3s in an effort to help tackle some of the challenges of deploying Kubernetes on the edge.
In this episode of TFiR: Let’s Talk, Michael Maxey, Vice President of Business Development at Zededa, talks about how the company is helping customers deploy workloads to edge devices. He talks about their key focus on small-footprint Kubernetes and how they are helping to tackle some of the challenges associated with this. Since the company has just announced a managed Kubernetes service, he also discusses the benefits of offering customers this option.
Key highlights of this video interview:
- Maxey gives us an oversight of Zededa, helping customers with edge orchestration and management. He explains that with their solution you can deploy workloads to edge devices running outside of the data center.
- Zededa has just announced a managed Kubernetes service. Maxey tells us that although their customers have been deploying Kubernetes on the edge for around three years, it was predominantly on a bring-you-own basis. He goes into detail about their new offering.
- Maxey discusses what sets Zededa apart from its competition saying they are bringing managed Kubernetes in a small footprint where customers have perhaps two or three devices at the edge.
- Maxey talks about the evolution of the edge market saying it is starting to accelerate. He takes us through some of the use cases they are now seeing more consistently: a couple of applications running on a device, which is running next to a Kubernetes cluster. He talks about how they are deploying full solutions around these sorts of use cases.
- For low-footprint Kubernetes use cases, Maxey tells us they typically use a lot of K3s. He explains some of the key features of this smaller distribution and how it helps with these sorts of cases.
- Maxey discusses the differences between full-scale Kubernetes and low-footprint Kubernetes saying it comes down to the size of the CPU and utilization. He talks about taking out a lot of the Kubernetes projects so that you end up running just the core services on the edge.
- Maxey explains the benefits of managed services on small footprint cases, saying they are partnering with SUSE to offer the managed service based on K3s. He explains it like having the cloud on your hardware at the edge and talks about how their managed service is helping customers so that they can focus on their applications.
- Kubernetes can be complex to manage and Maxey talks about the trends he is seeing saying they are seeing some customers using Kubernetes on the edge and others using different solutions. He feels that WebAssembly (Wasm) is showing potential to be a big player in the long term.
- With the evolution of Linux into becoming a significant open-source software collaboration, Maxey talks about whether he feels the same thing will happen with K3s in the Kubernetes space. He tells us that their EVE-OS is open source so anyone can come and contribute to it.
- Generative AI is a hot topic right now and Maxey explains that they are not seeing building big language models on the edge. However, he is seeing customers focus on data aggregation, tagging, and cleaning. Additionally, he is seeing customers take the LLM and do a video inference at the edge, or distributed machine learning.
- Maxey discusses the challenges he is seeing at the edge in the context of Kubernetes, saying the cost of Kubernetes workers visiting these remote sites can be expensive and how they need to bake in failover and failback because of this.
- Some of the use cases Zededa is seeing include computer vision like gas flares on an oil rig and the collection of data for starting to build an AI pipeline.
This summary was written by Emily Nicholls.
