Presentations
Modern cloud applications and platforms make it easy to use multiple data services and tools resulting in wide range of deployments, services and data stores within an organization.
This poses a challenge for data protection and disaster recovery because all those tools have their own mechanisms, datastores and recommendations for backup and restore.
In this talk we're going to learn how to use K8s tools to organize data protection processes and make day2 operations easier using CNCF project Kanister.
This session introduce a Dashboard-AI architecture running on Kata container
The platform is powered by LLM image to provide services such as data analysis, data forecast and dashboard generation
Usage of kata container allows us to scale and monitor a cluster of secured images
Get a technology overview of how Kata Containers, an open source, secure container runtime project, utilizes virtualization technology to overcome the biggest gap in traditional containers, the lack of isolation. The project is running in production in large public clouds, financial institutions, and being utilized for machine learning use cases to ensure data privacy. For an additional layer of protection Kata can be integrated with the Confidential Containers (CoCo) project, which brings confidential computing mechanisms into the container space.
Ever wanted your own private version of ChatGPT? LLMs offer powerful capabilities, but not everyone is comfortable sending their data over the internet to Microsoft, OpenAI, or Anthropic. Fortunately, open-source tools make it possible to set up and customize a local, secure LLM on your own terms. From installation to optimization, this talk will guide you through the process step by step, with demos to illustrate each stage. It’s time to build your own mini-Jarvis and start getting things done, efficiently, securely, and privately.
Replication conflicts can disrupt the consistency and reliability of a PostgreSQL database, especially in high-demand environments. This talk focuses on practical steps for monitoring, detecting, and resolving common replication conflicts.
Kernel level observability is critical for running resilient workloads. eBPF provides that visibility but can be challenging to learn. In this presentation we will introduce eBPF and demonstrate how to harness its superpowers for observability using the open source project Inspektor Gadget. Inspektor Gadget is a tool and framework for data collection and system inspection on Linux hosts and Kubernetes clusters using eBPF. We will show how its modular framework can collect real time metrics so that you have visibility into the Linux kernel in a customizable way.
Come get refreshed on the impact some basic choices in statistical behavior can have on what gets triggered. Learn why a median might be the choice for historical anomaly or sudden change. Jump into Gaussian distributions, data alignment challenges and the trouble with sampling. Walk out with a deeper understanding of your metrics and what they might tell you.
Running OpenStack on Kubernetes is a great idea for many reasons, but there is one very large drawback: when something goes wrong on Day 2, how do you find it? In this talk, we will explain our approach to a rather complex Observability problem using both OpenStack Services and Cloud Native Observability tools.
This session is for developers, operators and people interested OpenStack Kolla/Kolla-Ansible/Kayobe deployment projects.
It serves as an information gathering session for OpenStack Kolla developers to find out current/future users problems, requests for enhancements and discussion about project future plans.
It also serves as a venue for Kolla community to get together and grow the community in numbers, because that is what makes the open source projects stronger.
Journey into the realm of Kubernetes, where the Linux kernel and userspace tools sometimes clash, threatening your cluster's harmony. In this talk, we will dive into real-world stories of overcoming incompatibilities that can disrupt your clusters, from iptables quirks to modprobe mysteries. These challenges are not unique but are shared by the entire Kubernetes community. Together, we will explore strategies to tame these dragons and ensure that clusters thrive, with guidelines for harmonizing the Linux kernel and userspace tools.