Red Hat Leftovers
-
Red Hat Official ☛ Better understanding artificial intelligence and uncertainty
This article is written from the perspective of a Red Hat Technical Account Manager (TAM). A TAM’s unique role as the primary point of contact for a customer on Red Hat technical topics, combined with our advocacy for best practices, allows us to deeply understand customer needs and align those needs with Red Hat’s internal strategies. However, this does not resolve the prevalent uncertainty with AI/ML.
-
Red Hat Official ☛ Unlock your Automation Advantage with Red Hat Ansible Automation Platform 2.5
Ansible Automation Platform 2.5 is designed to make it easier than ever to create, deploy, and scale mission-critical IT automation workflows across the enterprise. With a reimagined platform experience, new capabilities, and key performance enhancements, this release is designed to help customers accelerate adoption, take on new use cases, and get more strategic value out of their automation initiatives. So, let’s dive in and talk about what’s new!
-
Red Hat ☛ Build container images in OpenShift using Podman as a GitLab runner
As you start your container journey, it is mandatory that you have image build tools. There are various tools available; the most common ones are Docker, Buildah, Podman, and Kaniko. You need to have image build tools to pull down images from external registry, customize the images, and build new images with libraries or applications. These images then published to container registries and further used in Kubernetes platforms such as Red Hat OpenShift, EKS, AKS, etc.
In this article, we'll be mainly talking about OpenShift, but the contents can be used for other platforms as well with little to no modifications.
In most enterprises, especially where security, compliance, and operational consistency are paramount, image build processes are strictly governed due to control risks. I have seen organizations having a separate build infrastructure just to manage the image build process. On the other side, few organizations allow application teams to manage build infrastructure. In both the cases, the infrastructure needs to have any of the above listed tools available. It is very common that such tools are mostly installed on virtual machines (VMs) reserved just for container build processes. Along with such infrastructure, it also comes with the overhead of maintenance, security, capacity management, and more.
-
Red Hat ☛ How to use RHEL as a WSL Podman machine [Ed: Red Hat now selling its main offering as... a Windows "app" that Microsoft controls]
This article from last year details the steps and actions required to build and run your Red Hat Enterprise Linux (RHEL) backdoored Windows Subsystem for GNU/Linux (WSL) image.
The purpose of this article is to describe the options needed for the RHEL WSL distribution so that it can be used as a Podman machine.
-
Red Hat ☛ How to fine-tune Llama 3.1 with Ray on OpenShift AI
This is the first in a series of articles that demonstrate the OpenShift Hey Hi (AI) tuning capabilities on a variety of Hey Hi (AI) accelerators. This post focuses on NVIDIA GPUs.
On July 23, 2024, Meta Hey Hi (AI) released Llama 3.1 models, their most capable LLMs to date, which are broadly available for a wide variety of use cases, from conversational assistants to agentic systems with APl calling capabilities. These models come in a herd of Llamas (pun intended) of three different sizes (8B, 70B, and 405B parameters), are trained on a corpus of about 15 trillion multilingual tokens, and provide an extended context window of up to 128 thousand tokens (a 250-page book, approximately).