Machine Learning

Open Data Hub 0.6.1: Bug fix release to smooth out redesign regressions

Open Data Hub 0.6.1: Bug fix release to smooth out redesign regressions

vpavlin
By Václav Pavlín

It is just a few short weeks since we released Open Data Hub (ODH) 0.6.0, bringing many changes to the underlying architecture and some new features. We found a few issues in this new version with the Kubeflow Operator and a few regressions that came in with the new JupyterHub updates. To make sure your experience with ODH 0.6 does not suffer because we wanted to release early, we offer a new (mostly) bugfix release: Open Data Hub 0.6.1.

Continue reading Open Data Hub 0.6.1: Bug fix release to smooth out redesign regressions

Share
Open Data Hub 0.6 brings component updates and Kubeflow architecture

Open Data Hub 0.6 brings component updates and Kubeflow architecture

vpavlin
By Václav Pavlín

Open Data Hub (ODH) is a blueprint for building an AI-as-a-service platform on Red Hat’s Kubernetes-based OpenShift 4.x. Version 0.6 of Open Data Hub comes with significant changes to the overall architecture as well as component updates and additions. In this article, we explore these changes.

From Ansible Operator to Kustomize

If you follow the Open Data Hub project closely, you might be aware that we have been working on a major design change for a few weeks now. Since we started working closer with the Kubeflow community to get Kubeflow running on OpenShift, we decided to leverage Kubeflow as the Open Data Hub upstream and adopt its deployment tools—namely KFdef manifests and Kustomize—for deployment manifest customization.

Continue reading “Open Data Hub 0.6 brings component updates and Kubeflow architecture”

Share
AI/ML pipelines using Open Data Hub and Kubeflow on Red Hat OpenShift

AI/ML pipelines using Open Data Hub and Kubeflow on Red Hat OpenShift

nakfour
By Juana Nakfour

When it comes to the process of optimizing a production-level artificial intelligence/machine learning (AI/ML) process, workflows and pipelines are an integral part of this effort. Pipelines are used to create workflows that are repeatable, automated, customizable, and intelligent.

An example AI/ML pipeline is presented in Figure 1, where functionalities such as data extract, transform, and load (ETL), model training, model evaluation, and model serving are automated as part of the pipeline.

Continue reading “AI/ML pipelines using Open Data Hub and Kubeflow on Red Hat OpenShift”

Share
Microbenchmarks for AI applications using Red Hat OpenShift on PSI in project Thoth

Microbenchmarks for AI applications using Red Hat OpenShift on PSI in project Thoth

fmurdaca
By Francesco Murdaca

Project Thoth is an artificial intelligence (AI) R&D Red Hat research project as part of the Office of the CTO and the AI Center of Excellence (CoE). This project aims to build a knowledge graph and a recommendation system for application stacks based on the collected knowledge, such as machine learning (ML) applications that rely on popular open source ML frameworks and libraries (TensorFlow, PyTorch, MXNet, etc.). In this article, we examine the potential of project Thoth’s infrastructure running in Red Hat Openshift and explore how it can collect performance observations.

Several types of observations are gathered from various domains (like build time, run time and performance, and application binary interfaces (ABI)). These observations are collected through the Thoth system and enrich the knowledge graph automatically. The knowledge graph is then used to learn from the observations. Project Thoth architecture requires multi-namespace deployment in an OpenShift environment, which is run on PnT DevOps Shared Infrastructure (PSI), a shared multi-tenant OpenShift cluster.

Continue reading “Microbenchmarks for AI applications using Red Hat OpenShift on PSI in project Thoth”

Share
Python wheels, AI/ML, and ABI compatibility

Python wheels, AI/ML, and ABI compatibility

smodeelcodonell
By Subin Modeel and Carlos O'Donell

Python has become a popular programming language in the AI/ML world. Projects like TensorFlow and PyTorch have Python bindings as the primary interface used by data scientists to write machine learning code. However, distributing AI/ML-related Python packages and ensuring application binary interface (ABI) compatibility between various Python packages and system libraries presents a unique set of challenges.

The manylinux standard (e.g., manylinux2014) for Python wheels provides a practical solution to these challenges, but it also introduces new challenges that the Python community and developers need to consider. Before we delve into these additional challenges, we’ll briefly look at the Python ecosystem for packaging and distribution.

Continue reading “Python wheels, AI/ML, and ABI compatibility”

Share
Red Hat Summit 2019 Labs: Emerging technology roadmap

Red Hat Summit 2019 Labs: Emerging technology roadmap

ericschabell
By Eric D. Schabell

Red Hat Summit 2019 is rocking Boston, MA, May 7-9 in the Boston Convention and Exhibition Center. Everything you need to know about the current state of open source enterprise-ready software can be found at this event. You’ll find customers talking about their experiences leveraging open source in their solutions, creators of open source technologies you’re using, and hands-on lab experiences relating to these technologies.

This hands-on appeal is what this series of articles is about. In previous articles, we looked at labs focusing on Red Hat Enterprise Linux, Integration and APIs, and cloud-native app development. In this article, we’ll look at labs in the “Emerging Technology” track.

Continue reading “Red Hat Summit 2019 Labs: Emerging technology roadmap”

Share