Presentations
privacyIDEA is a multi factor authentication system that helps administrators, organizsations and companies to manage thousands of 2nd factors of users. privacyIDEA supports all different token types and runs on premies. The project is around for 10 years. While it is completely open source, a company run by the developers provides support to companies who require this. This proves that open source can be the base for a successful business model. We will talk about how privacyIDEA developed, what are the future plans and how it can also help you solve your authentication challenges.
K8s’ frequent pod kills, and evictions make it challenging to operate non-fault-tolerant stateful applications that were never designed for frequent cold restarts. In this talk, we will describe the technology and use cases for a Kubernetes operator that addresses these operational challenges by enabling stateful applications to hot restart from where they were interrupted. The underlying technology will be demonstrated for CPU-only and GPU-accelerated stateful applications.
PostgreSQL 17 is almost at feature freeze, and at this point we have at least some ideas of what features will be included. This talk will take a quick look at what's in there already, and what's in the queue for the next major release!
This session will contain an introduction to Nix for the uninitiated. You will leave with an understanding of the basics and, perhaps, the beginning of a lifetime of love for Nix. We will cover installation, use of the CLI, and the basics of the Nix language.
What do you do once you’ve bumped up against one or more of Prometheus' limitations?
- Outgrown what a single instance can handle
- Many separate instances and need a global view
- Need longer retention of your metric series
- Require multi-tenancy
- Need better HA/resiliency
The classic “It Depends.” applies here.
Old, incorrect or false information online can make this even harder.
Come step through the popular OSS options for scaling Prometheus, and gain guidance on the next stage of your observability journey!
In the intricate landscape of software development and system analysis, tools like readelf, nm, and objdump have been the go-to solutions for delving into Executable and Linkable Format (ELF) objects. These tools, while robust, follow the pattern of many Unix utilities and dump unstructured output to the console which can make it difficult to perform advanced analysis.
We present 'sqlelf' which allows declarative access to ELF files. It brings a refreshing shift, offering a friendlier and more versatile way to explore ELF objects using SQL.
In this talk, you will learn about the Ubuntu HPC community team’s effort to bridge that skill gap and accelerate users’ time to science on Ubuntu. Like how Ubuntu revolutionised the Linux desktop world in 2004, the Ubuntu HPC community team is aiming to revolutionise HPC by developing an open source supercomputing infrastructure stack that anyone can freely use.
The presentation covers using Linux on small board computers or Industrial IOT machines to run inference on models.
The speaker will present some out of the box solutions such as the Up Squared, Raspberry Pi and more.
The talk will cover how to install deep learning solutions and use pre-trained models and running inference.
This talk does not cover training your own model.
The audience will be presented with information about using AI on Linux on small board and industrial AIOT machines.
Trusting a friend and trusting a service are fundamentally different, and current AI systems are poised to exploit that difference. We will think of them as friends when they are actually services, and confidents when they will be actually collecting data behind our backs. Government is how we create social trust in our society, which requires regulating this potentially intimate technology.