Name: From GPU Chaos to AI Factory: How Enterprises are Building Repeatable AI Pipelines at Scale
Start: 2025-08-20T16:00:00Z
End: 2025-08-20T16:01:00.000Z
Location: BrightTALK
Rating: 0

Mirantis helps organizations ship code faster utilizing public and private clouds, providing a public cloud experience on any infrastructure from the data center to the edge. With a long-standing record of delivering solutions based on open source with enterprise-grade support, Mirantis offers an as-a-service experience of Kubernetes and OpenStack to provide containerization and virtualization services. Mirantis serves many of the world’s leading enterprises, including Adobe, DocuSign, Liberty Mutual, Nationwide Insurance, Reliance Jio, S&P Global, Seagate, Société Générale, and Volkswagen. Learn more at www.mirantis.com.

Mirantis is a global company headquartered in Campbell, California.

Sovereign cloud providers and regional telcos face mounting pressure to deliver scalable, profitable AI infrastructure without the complexity or cost of hyperscaler stacks. Traditional IaaS models were not designed for the elasticity, scale, or GPU efficiency that modern AI workloads demand. As a result, many providers experience sluggish customer onboarding, idle GPU capacity, and missed monetization opportunities.

In this webinar, you will learn how to turn your GPU infrastructure into a high-margin AI service platform. Discover how providers are using k0rdent AI’s turnkey deployment templates to launch GPU-backed services in days, reduce time to market, optimize GPU usage, and eliminate infrastructure waste.

To bring the ROI story to life, we will showcase live OpenCost-Grafana dashboards and forecasting metrics that highlight cost efficiency, revenue impact, and long-term profitability. You will walk away with practical insights to help you build a scalable and sovereign AI cloud strategy.

What you’ll learn:
• Why traditional cloud infrastructure falls short for AI workloads and what to do about it
• How to accelerate customer onboarding with k0rdent AI’s GPU service templates
• How to maximize GPU utilization and forecast ROI with OpenCost-Grafana dashboards

This session is ideal for Heads of Cloud Services, Product VPs, and CTOs, and also valuable for anyone involved in building AI infrastructure or launching GPU-backed services in sovereign, regulated, or regional markets.

How Sovereign Cloud Providers Can Monetize GPU Infrastructure Without Hyperscaler Complexity

Scaling AI across edge, core, and cloud environments is now a top priority but also a top challenge. How do you deliver high-performance inference, maintain control over your data, and keep costs predictable across a hybrid or multi-cloud setup?

In this session, you'll learn how to design and operationalize GPU-powered infrastructure that meets enterprise-grade AI demands. Whether you're leading platform architecture or responsible for AI/ML infrastructure strategy, you'll walk away with:
• Proven patterns for building hybrid GPU clouds that support both training and inference
• A framework for balancing performance, cost, and data sovereignty
• Strategies to enable inference-as-a-service across teams and regions
• Insights into how Kubernetes-native orchestration and NVIDIA technologies can accelerate deployment at scale

Led by Senior DevOps Engineer Prashant Ramhit, this session offers actionable guidance based on real-world deployments. If you're building or scaling intelligent applications, this is a can't-miss discussion on how to do it right.

AI at the Edge, Core, and Cloud: Building GPU Clouds for the Next Era of Intelligent Applications

As AI adoption accelerates, sovereign cloud providers and regulated enterprises face increasing pressure to offer scalable AI services while maintaining full control over data, infrastructure, and compliance. However, building secure, multi-tenant GPU infrastructure across hybrid environments remains a major operational challenge.

In this session, you’ll learn how k0rdent AI helps organizations streamline GPU provisioning, and deploy AI workloads at scale. The platform is built on a Kubernetes-native architecture designed for high-compute, security-sensitive environments.

We’ll also explore a real-world case study featuring Nebul, a sovereign cloud provider based in the Netherlands, and show how they operationalized their GPU clusters to serve sensitive AI workloads securely and efficiently.

What you’ll learn:
- How to automate GPU-ready Kubernetes clusters with compliance in mind
- Best practices for deploying and scaling multi-tenant AI inference services using KServe, Knative, and Istio
- Lessons from Nebul’s journey as a sovereign CSP scaling secure AI infrastructure

Whether you’re building sovereign cloud services or modernizing infrastructure for sensitive AI workloads, this session will give you the blueprint to do it securely, efficiently, and at scale.

Sovereign AI Infrastructure at Scale: Real-World Blueprint for Cloud Providers

Managing Kubernetes clusters across cloud, on-prem, and edge environments introduces challenges in scalability, networking, and security.

In this joint webinar, Mirantis and Calico experts  will demonstrate how the open source tools k0rdent and Calico work together to simplify multi-cluster operations and enhance network visibility and protection.

You’ll learn how Mirantis k0rdent makes it easier to provision and manage Kubernetes clusters across AWS and Azure, with Calico automatically deployed as the CNI. k0rdent simplifies multi-cluster operations by centralizing lifecycle management and policy enforcement, helping teams maintain consistency across diverse environments with less manual effort.

You’ll also explore how Calico provides deep visibility into Kubernetes network traffic, strengthens network security, and improves overall reliability. The session will cover how Calico supports scalable, secure networking—and how organizations with more advanced needs can extend to Calico Enterprise for unified network security management across hybrid and multi-cloud deployments.

Whether you're running a few clusters or operating at global scale, this session will share practical strategies to simplify operations and strengthen your Kubernetes networking posture.

Simplify and Secure Multi-Cluster Kubernetes with k0rdent and Calico

As Kubernetes environments grow across clouds and data centers, ensuring consistent security, compliance, and operational policies becomes critical — and increasingly complex.

Join Mirantis and Nirmata to see how open source k0rdent and Kyverno make policy-driven Kubernetes management simple, scalable, and automated across multi-cluster environments.

In this session, you’ll discover how:
- Discover how k0rdent simplifies creating and managing secure multi-cluster Kubernetes environments across clouds and data centers.
- Learn how Kyverno automates policy-driven governance, compliance enforcement, and security across distributed Kubernetes clusters.
- See how k0rdent and Kyverno together simplify operations, enable secure developer self-service, and accelerate cloud-native initiatives.

Whether you're building internal developer platforms (IDPs), supporting AI-driven workloads, or securing large-scale Kubernetes deployments, this webinar will show you how to deliver secure, compliant, and scalable Kubernetes environments — with less complexity and more control.

Policy-Driven Multi-Cluster Kubernetes for AI Workloads Made Simple with k0rdent and Kyverno

Deploying AI on Kubernetes can be complex, involving GPU provisioning, model serving, and cost-efficient scaling. This session explores how to streamline AI deployments on Kubernetes using k0rdent. 

Attendees will learn how to automate GPU ready cluster provisioning, deploy scalable AI models with Kserve, and optimize inference workloads with Knative and Istio. With GPUs being constrained and costly, this session emphasizes how auto-scaling can maximize resource utilization while minimizing operational costs.

The talk will also cover GPU monitoring strategies using Prometheus and Grafana to ensure optimal performance. Join us to see how k0rdent empowers AI teams to deploy, scale, and optimize models effortlessly—ensuring maximum performance at the lowest cost.

AI on Kubernetes: Scaling Smarter, Running Faster with k0rdent

As financial institutions modernize their IT infrastructure, traditional virtualization solutions often impose high costs, operational complexity, and vendor lock-in. Cloud-native virtualization offers a scalable, cost-efficient alternative that enhances agility, security, and compliance—critical factors for financial services innovation.

In this webinar, we’ll explore a real-world use case of cloud-native virtualization in the financial sector. Our speaker, Michael Tillman, Solution Architect Manager, will discuss how a leading financial institution successfully transitioned from legacy virtualization to a Kubernetes-powered, cloud-native approach. 

You'll gain insights into:
- The challenges of traditional virtualization in financial services
- How cloud-native virtualization optimizes performance and reduces costs
- Security, compliance, and operational benefits for regulated environments
- Lessons learned and best practices for IT teams in banking and financial services

Join us to discover how financial institutions can achieve greater flexibility and efficiency with cloud-native virtualization. Don't miss this opportunity to learn from real-world experiences and get your questions answered by cloud native experts!

From Legacy to Cloud-Native: Virtualization Strategies for Financial Institutions

For SRE or DevOps professionals navigating uncharted waters of LLMs, RAGs and GenAI, the lack of standardization, resources and knowledge can quickly turn into an operational nightmare. DevOps and SRE practices are no longer just about infrastructure and delivery speed — they now also delve into the complexities of GenAI, particularly LLMs. 

While it may be tempting for an uninitiated SRE to approach this as a traditional MLOps problem, it quickly becomes clear that LLMOps is an entirely different challenge, starting from day 0. 

Join our webinar led by Harsh Mishra, Site Reliability Engineer at One2N Consulting, to explore the unique challenges and solutions of LLMOps from an SRE perspective.

What we’ll cover:
- Day 0 prerequisites for understanding and hosting LLMs on Kubernetes.
- Breaking down core components: AI (RAGs, Vector Databases), Backend, and Compute.
- Key differences between LLMOps and traditional MLOps.
- Case study: Building an internal tool with optimized architecture using Nvidia GPUs and Ray Distributed.
- Lessons learned in our journey in helping organisations build the necessary GenAI muscle for SREs.

Register now!

The SRE’s Guide to LLMOps: Day-0 Deploying and Managing AI Workloads on Kubernetes

Managing Kubernetes at scale can be complex, with operational overhead and vendor lock-in limiting flexibility and innovation. Open-source solutions like k0rdent empower enterprises with a vendor-neutral, scalable, and efficient approach to Kubernetes management.

Join us for an insightful webinar where we explore how k0rdent, an open-source Distributed Container Management Environment (DCME), provides a vendor-neutral, scalable, and efficient approach to multi-cluster Kubernetes management—giving platform engineers and IT leaders the control, automation, and flexibility they need.

What You’ll Learn:
- Break Free from Vendor Lock-In: Discover strategies to build a flexible, open Kubernetes ecosystem.
- Simplify Multi-Cluster Management: Step-by-step guidance on managing Kubernetes at scale with reduced complexity.
- Optimize Workloads for Efficiency: Real-world use cases of organizations leveraging k0rdent to enhance performance and lower costs.
- Seamless Integration with Existing Operations: Actionable insights on incorporating k0rdent into your infrastructure for improved automation and scalability.

Whether you’re a platform engineer, DevOps leader, or IT decision-maker, this session will provide practical knowledge to elevate your Kubernetes strategy.

Don’t miss this opportunity to take control of your Kubernetes environment with open-source innovation. Register now and take the next step toward a more agile, scalable future with k0rdent.

The Case for Open Source: How k0rdent Empowers Enterprise Agility

Modern governments face growing demands to modernize IT systems, improve scalability, and deliver secure, efficient services. Cloud-native virtualization offers a streamlined path to transformation, enabling governments to overcome challenges and build future-ready infrastructure.

Join us for a webinar as we explore a real-world use case that demonstrates the benefits of adopting cloud-native technologies in a government setting.

You’ll learn:
- How cloud-native virtualization reduced operational complexity.
- Strategies to achieve scalability and improve service efficiency.
- Real-world insights into overcoming legacy system challenges.
- Best practices for implementing secure, future-ready infrastructure.

Gain actionable insights into overcoming modernization challenges and building future-ready infrastructure solutions.

Cloud Native Virtualization in Government: A Real-World Use Case

Kubernetes platform engineering is complex, especially with large, shared clusters supporting multiple workloads and teams.

Defeating complexity requires multi-cluster thinking: Compose clusters declaratively with services and policies to support one or a small range of workload types (versus many); administer and lifecycle manage declaratively to continually assure cluster state; and repeat for all the workload types your organization needs to support.

Simplifying platforms and customizing them to workload requirements pays off bigtime in agility, smoother operations, higher workload availability, security, and compliance. Plus: You don't need to implement a global multi-cluster architecture to do this at smaller scales.

In this webinar, we’ll explore:
- How to think of Kubernetes platforms as containers for workloads, services (LMA, FinOps), and policies.
- Building up use-case-optimized platforms from 'bare clusters.'
- Approaches to automate platform operations and LCM.
- Strategies to keep platforms healthy and integrated over time.

And we'll share how we're doing all of this with open-source-based tools. Join us!

Defeat Complexity with Use-Case-Optimized Kubernetes Platforms

In this webinar, we'll delve into the journey of a SaaS company that slashed costs and boosted efficiency by migrating 3,000 VMs to OpenStack with CI/CD automation. The company faced high costs and scalability limitations with VMware, prompting the shift to a more flexible and cost-effective OpenStack environment. Mirantis played a key role in this transition, swiftly designing and deploying a FedRAMP High-compliant on-premises cloud infrastructure within just six months.

This session will cover:
- Strategy and automation workflow details for rapidly migrating 3,000 VMs.
- Insights into how Mirantis designed and implemented a secure, compliant cloud infrastructure so quickly.

Join us to discover how strategic cloud migrations and compliance frameworks can reduce costs and improve scalability, offering valuable insights for your organization’s transformation.

How a SaaS Company Saved 60% by Migrating from VMware

Docker. Kubernetes. Dix ans déjà que ces technologies révolutionnent notre industrie, mais nombreux sont les utilisateurs qui se confrontent à des soucis d'échelle

Dès lors que nous utilisons des clusters mutualisés, les soucis apparaissent : sécurité, cycle de vie, dépendances…

L’intérêt principal de cette technologie est sa versatilité. Après le déploiement de composants, le déploiement d'applications complexes, la mise à disposition de namespaces dédiés, voilà que Kubernetes permet… la mise à disposition de clusters Kubernetes à la demande. Dans son datacenter, chez différents hébergeurs ou acteurs du Cloud public.

Nous allons voir comment mettre cela à disposition de vos projets, via ClusterAPI et k0smotron, tout en conservant la maîtrise de votre flotte de clusters. Nous verrons aussi comment fournir cela en répondant à des standards d'entreprise en termes de classe de service, de sécurité d'accès, de sauvegarde et d'observabilité.

Déployer en self-service des clusters Kubernetes adaptés à vos standards

As Kubernetes dominates container orchestration, its complexity and resource demands drive interest in lighter alternatives. This talk explores k0s, a CNCF-compliant Kubernetes distribution celebrated for its lightweight, easy-to-maintain design. We will examine k0s's architecture, emphasizing its minimal dependencies and simplified management, which includes a single-binary setup process.

Comparing k0s to other distributions like k3s and minikube, we highlight its advantages across various deployment scenarios, from bare-metal to the cloud. We'll demonstrate why k0s offers a simpler, yet functional approach to Kubernetes adoption.

Additionally, we introduce k0smotron, which enables the creation of a Kubernetes control plane within a cluster, facilitating Kubernetes as a Service. This feature underscores k0s's capability to enhance scalability and management in diverse environments.

k0s: Yet Another Lightweight Kubernetes Distribution?

As IT departments face economic pressures and the uncertainty surrounding VMware, many are turning to OpenStack as a cost-effective, future-proof alternative for on-premises infrastructure.

In this session, Christian Huebner, Mirantis Director of Architecture Services, and Heiko Kraemer, Principal Cloud Architect, will explore how transitioning from VMware to OpenStack can reduce costs and eliminate vendor lock-in. Together, they will guide you through the complexities of working with OpenStack’s core components—compute, software-defined storage (SDS), and software-defined networking (SDN)—demonstrating how to repurpose existing hardware to minimize upfront capital expenditure and accelerate return on investment (ROI).

Christian and Heiko will also cover cloud planning and deployment strategies that simplify OpenStack adoption, alongside best practices for monitoring and life cycle management (LCM) to ensure seamless Day 2 operations and optimized workload performance. Additionally, they will provide proven methodologies for migrating VMware workloads to OpenStack, ensuring operational continuity while enhancing agility and efficiency.

Attend this webinar to compare VMware and OpenStack capabilities in detail, and learn:
- Detailed comparison between VMware and OpenStack capabilities
- How to repurpose hardware for faster ROI with minimal investment
- Best practices for simplifying deployment and optimizing OpenStack operations
- Proven methods for migrating VMware workloads to boost agility and performance

By the end of this session, you’ll have gained a comprehensive understanding of how OpenStack delivers significant business value, from substantial cost savings to increased control and scalability.

Why OpenStack is a Viable Alternative to VMware

Certain VM workloads are crucial and not obsolete, but lack a clear path to modernization. Currently, they run on IaaS or public clouds, which are costly and require time-intensive maintenance from operators and developers.

What if you could run these VMs on Kubernetes, inside pods, alongside containers, and do so very affordably – i.e., for much less than VMware or AWS?

With MKE 4 and MKE Virtualization (KubeVirt) you can easily move VM workloads to Kubernetes and gain:
- Kubernetes-native high availability and autoscaling for VMs
- Simplified ways to extend VM functionality with adjacent containerized services
- Enhanced security with consistent policy application
- Automation through GitOps and Kubernetes-native configuration
- A pathway to modernize VMs into scalable, containerized apps at your own pace

In this webinar, we’ll cover the economics, principles, and strategies for selecting workloads suited to this approach.

Migrating VMs to Kubernetes: Is It a Practical Solution?

Depuis plus de dix ans, OpenStack s’est imposé comme le principal framework open source pour les clouds privés, prouvant son efficacité à l’échelle internationale.
De nombreuses organisations exploitent OpenStack en l’adaptant à leurs propres besoins mais, bien qu'OpenStack soit particulièrement performant dans l'Infrastructure as a Service (IaaS), ses fonctions principales sont centrées autour de l'orchestration flexible et ouverte du Calcul, du Réseau et du Stockage. Ce positionnement laisse à désirer pour les organisations cherchant des solutions de Platform as a Service (PaaS), un manque généralement comblé par l’implémentation de cloud public, qui offre des services intégrés tels que Database as a Service (DBaaS) et Message Queue as a Service (MQaaS).

Chez Mirantis, nous répondons à tous les besoins de nos clients grâce à des solutions open source complètes, permettant aux entreprises de bénéficier d’une véritable alternative aux offres partielles.

Durant notre prochain webinaire, Daniel Virassamy expliquera comment OpenStack prend en charge les fonctionnalités PaaS, offrant ainsi une plateforme complète pour faciliter le développement de logiciels et l'hébergement en production. Le webinaire comprendra :
- Un aperçu complet d’une architecture couvrant à la fois les couches IaaS et PaaS.
- Une vue détaillée du Software Bill of Material (SBOM) et du Bill of Material (BOM) pour une solution complète basée sur l'open source, garantissant ainsi transparence et efficacité.
- Les services principaux fournis par la couche IaaS, facilitant la gestion de l'infrastructure.
- L'intégration poussée du PaaS avec l'IaaS, assurant une interopérabilité fluide et des fonctionnalités améliorées.

À travers nos initiatives, nous permettons aux entreprises d'adopter pleinement les solutions open source, leur offrant facilité, évolutivité et rentabilité dans leurs déploiements cloud.

Libérer la puissance de l'open source : le cloud privé en tant qu’alternative au cloud public

AI initiatives often start with promise, but scaling them into production is where most organizations struggle. The result is fragmented teams, underutilized GPU clusters, rising costs, and ad hoc AI projects that cannot be repeated or sustained.

In this webinar, we will explore how forward-thinking enterprises are transforming GPU sprawl into highly efficient AI Factories by building repeatable, scalable AI and ML pipelines. You will learn how to:
• Streamline GPU provisioning and orchestration across hybrid and multi-cloud environments
• Establish standardized, reusable workflows for training, inference, and deployment
• Optimize infrastructure utilization to eliminate bottlenecks and control costs
• Enable platform teams to deliver self-service AI environments for data scientists and developers

This session will provide a blueprint for taking your AI strategy from disorganized experimentation to an industrialized factory model that accelerates innovation and delivers measurable ROI at enterprise scale.

Who should attend: Platform engineers, AI and ML architects, DevOps leaders, and technology decision-makers responsible for enterprise AI adoption

From GPU Chaos to AI Factory: How Enterprises are Building Repeatable AI Pipelines at Scale

GPU Cloud

AI Factory

AI Cloud

AI Infrastructure

AI/ML

Welcome to the virtualization community on BrightTALK! Whether it affects servers, storage, networks, desktops or other parts of the data center, virtualization provides real benefits by reducing the resources needed for your
infrastructure and creating software-defined data center components. However, it can also complicate your infrastructure. Join this active community to learn best
practices for avoiding virtual machine sprawl and other common virtualization pitfalls as well as how you can make the most of your virtualization environment.

Virtualization

Cloud computing has exploded over the past few years, delivering a previously unimagined level of workplace mobility and flexibility. The cloud computing community on BrightTALK is made up of thousands of engaged professionals learning from the latest cloud computing research and resources. Join the community to expand your cloud computing knowledge and have your questions answered in live sessions with industry experts and vendor representatives.

Cloud Computing

The IT security community on BrightTALK is composed of more than 200,000 IT security professionals trading relevant information on software assurance, network security and mobile security. Join the conversation by watching on-demand and live information security webinars and asking questions of experts and industry leaders.

IT Security

The data center management community focuses on the holistic management and optimization of the data center. From technologies such as virtualization and cloud computing to data center design, colocation, energy efficiency and monitoring, the BrightTALK data center management community provides the most up-to-date and engaging content from industry experts to better your infrastructure and operations. Engage with a community of your peers and industry experts by asking questions, rating presentations and participating in polls during webinars, all while you gain insight that will help you transform your infrastructure into a next generation data center.

Data Center Management

The application development community features top thought leadership focusing on optimal practices in software development, SDLC methodology, mobile app development and application development platforms and tools. Join top software engineers and coders as they cover emerging trends in everything from enterprise app development to developing for mobile platforms such as Android and iOS.

Application Development

Network infrastructure professionals understand that a reliable and secure infrastructure is crucial to enabling business execution. Join the network infrastructure community to interact with thousands of IT professionals. Browse hundreds of on-demand and live webinars and videos to learn about the latest trends in network computing, SDN, WAN optimization and more.

Network Infrastructure

As an IT professional, many of the problems you face are multifaceted, complex and don’t lend themselves to simple solutions. The information technology community features useful and free information technology resources. Join to browse thousands of videos and webinars on ITIL best practices, IT security strategy and more presented by leading CTOs, CIOs and other technology experts.

From GPU Chaos to AI Factory: How Enterprises are Building Repeatable AI Pipelines at Scale

Presented by

About this talk

Mirantis