Welcome to Cloud Native Trento (& now Bolzano/Bozen)!

Join us for this event and help us shape the world of cloud native technologies in Trentino Alto-Adige.

We're an official user community group of the Cloud Native Computing Foundation (CNCF).

What to expect

Vendor-agnostic talks, no sales pitch, from practitioners to practitioners

• 🗣️ Two or Three speakers

• 👥 Meet old and new friends from the community

• 🍕 Free food and drinks afterwards

• ✔️ Just RSVP and book your ticket (it's free, but we need to know how much food and drinks to buy)!

Agenda

• 17:30 Doors open and welcome (intro and talks start at 18:00)

• 18:00 Cloud Native Trento Organizers - Kubecon & CloudNativeCon 2026 recap: updates from the community

• 18:20 Philip Laine - Cache me if you can: P2P Image Sharing in Kubernetes

• 19:10 Davide Sbetti - LLM-Aware Load Balancing on Kubernetes

• 19:30 Food and drinks

Speakers

Philip Laine is a senior platform engineer and open-source maintainer. He runs Spegel, a stateless cluster-local OCI registry mirror for Kubernetes, and is a former core maintainer of Flux.

Davide Sbetti is a software engineer at Würth IT Italy; he is passionate about Data Science and Machine Learning.

Sessions

We will start with an intro to CNCF and a recap of the latest news and trends from Kubecon and CloudNativeCon 2026.

1. Cache me if you can: P2P Image Sharing in Kubernetes

External image registries can be slow, inefficient, and prone to network issues or down time leading to failure to meet SLAs or even worse outages. It’s time to explore different strategies for creating more resilient image registries. We will explore different image caching approaches such as Harbor, which can be leveraged as a pull through mirror. Next, we will look into how to preload images on the machine images themselves. Lastly, we’ll weigh the benefits and costs of solutions in the eco-system such as Spegel, Kraken, or Dragonfly. Attendees will learn to deploy these solutions to optimize their Kubernetes workloads, decrease pod startup times, and improve the performance of edge node deployments in their own production clusters.

2. LLM Aware Load Balancing on Kubernetes

During this talk we will explore some possibilities to manage LLM workloads on Kubernetes, with a particular analysis of load balancing techniques.

Transportation

Need a ride from Trento or want to offer one? Just reach out to cloudnativetrento@gmail.com or on LinkedIn

Parking

You will find parking along the road in the vicinity of the office (Wuerth IT Italy S.R.L., 4 Via Johann Kravogl, Bolzano). If those are full, you can still find other options within 10-15m walking distance on Google Maps (e.g. https://maps.app.goo.gl/UFSw6un5pAskt7sk6, https://maps.app.goo.gl/yTJTUubFXpmMEPhdA, https://maps.app.goo.gl/sJYTNYjVJmsqhHmE6).

Our Sponsor: Würth IT Italy

Würth IT Italy delivers global IT expertise, combining innovation, security, and reliability to support business-critical systems worldwide. With an international footprint and a future-oriented mindset, the company helps shape modern digital environments.

Check them out at https://www.wuerth-it.it/en/

🌐 Stay connected with the community on LinkedIn or join the conversation in the CNCF Slack at #cloud-native-trento.https://www.linkedin.com/in/phillebaba/