Keith Tobin

#Snowflake introduces new feature Workload Identity Federation — a new way to authenticate 🔐 without managing passwords or key pairs. With this feature, services running on #AWS (#Azure, #GCP)compute (#ec2, #Lambda, ECS, EKS, etc.) can authenticate directly to Snowflake using their native cloud identity, eliminating the need for Password rotation, Public/private key management, and extra credential storage. This enables secure, scalable, and cloud-native service-to-service authentication, simplifying DevOps and reducing security risks. Check it out https://lnkd.in/gXBwWvkF

262

12 Comments

Open Hybrid Cloud, PoT use cases and recorded demos. In the previous post "What a busy week!" - https://lnkd.in/eESR9rQf - we have shared the list of use cases along with the link to the Github repo where all PoT assets are maintained. This slide deck provides a short overview of each use case with a link to a recorded demo. The goal is to make it easier to understand the context behind each use case and see it in action. You can also see the products used in the demo in addition to target personas. NOTE: You probably need to download the PDF for the links to work. You can also find all recorded demo on below Youtube playlist https://lnkd.in/e__XKJdS #redhat #openshift #open_hybrid_cloud

104

6 Comments

First Cilium became the default for every major cloud provider’s Kubernetes distribution. Now it’s the first and only CNI certified across the entire Red Hat ecosystem (Network Conformance, OpenShift Virtualization, OpenShift Service Mesh, and OpenShift Hosted Control Planes). Cilium isn’t just the community’s favorite CNI, it’s becoming the standard for Kubernetes networking everywhere it runs. No matter where enterprises need Kubernetes, Cilium has become the default networking layer that they can rely upon. Home lab, cloud, on premise, what else is left for Cilium to become the standard? https://lnkd.in/dfWkiZPg

134

3 Comments

\o/ Today, we’re announcing Amazon ECS Managed Instances, a new compute option for Amazon Elastic Container Service (Amazon ECS) that enables developers to use the full range of Amazon Elastic Compute Cloud (Amazon EC2) capabilities while offloading infrastructure management responsibilities to Amazon Web Service (AWS). This new offering combines the operational simplicity of offloading infrastructure with the flexibility and control of Amazon EC2, which means customers can focus on building applications that drive innovation, while reducing total cost of ownership (TCO) and maintaining AWS best practices. https://lnkd.in/dJYtU9vw

86

2 Comments

The hyperscalar clouds need a proprietary network to build an AZ, Nutanix figured out how to do it in software using a clustered IO path and a stateless hypervisor. For 20 years, I've seen distributed systems struggle with a fundamental problem: making infrastructure truly transactional. Before the hyperscalar cloud, we had transactional infrastructure, but the cloud forced us to reconsider our design. Following the cloud, the industry relied on people and observability to simulate transactional behavior. If an operation failed, we spent people, time, and money to figure out why. My own journey led to a key realization: the industry lacked a clear definition of an "Availability Zone" (AZ). A true AZ is a control plane that, if it fails, makes the hardware it manages unusable. This is why cloud providers have an advantage—they engineer their entire control plane around this principle. How did they get there? The hyperscalar cloud underlay network is a thing of magic. What they did was guarantee that if a host was unreachable, it was effectively down. And then they built all of their system software to apply that principle. But what about on-premises environments? On-prem cannot rely on such a magical network, for many reasons. And because they can't, all hell breaks loose. Because the network doesn't enforce the rigid rules I described, unreachability only means you can't reach the system. There is no guarantee that the workload isn't running. As a result, any operation on the machine where the workload is running is fraught with peril. That's where Nutanix comes in. They've found a way to deliver transactional, cloud-like behavior without a proprietary network. Their key innovation? A clustered IO data path and a stateless hypervisor that ensures the hypervisor can only access state that the control plane allows. The clustered IO path guarantees that the state of the system cannot change behind the back of the cluster. The stateless hypervisor ensures that only workloads authorized by the clustered control plane are running. Once you build that IO path, you can then make an AZ that is deterministic. Why, much like the cloud exploits their network design to know what is up and down and connected to what , we can exploit the fact that we know which workload is running, and we know which workload is not. We always see the state of the system. It's a game-changer for on-prem environments and a powerful example of how solving the right problem—not just the obvious one—can lead to true innovation. I've benefited from a lot of different people over the years in my journey, but one of them was the incredible Dahlia Malkhi, who is currently a professor at UCSB. You can find a longer version here. https://lnkd.in/gZ88DqfD #distributedsystems #cloudcomputing #datacenter #infrastructure #nutanix

31

NEW ANALYST BRIEF: Oracle’s distributed database on Exascale aims to support the era of agentic AI, hyperscale OLTP, and data residency. It combines raft-based replication, zero data loss, Exadata performance, and a serverless cloud model on OCI to deliver a powerful, elastic data platform that enterprises can deploy globally. Full Analysis link below 👇 “For organizations to be successful with agentic AI, they need a nearly infinite data platform architecture that helps support data residency, and Oracle has delivered that with its latest release. Oracle Distributed Database on Exascale leverages Exadata’s exceptional high availability and AI processing performance to support critical back-end systems executing agent-generated tasks. AI and Agents will be distributed, so your data must also be. At the same time, Exascale’s hyper-elastic technology with pay-per-use capabilities ensures that cost efficiency is very compelling. This is a game changer for organizations looking for the latest in distributed database technology,” — Rob Strechay, Managing Director & Principal Analyst, theCUBE Research. Oracle #Exadata #Exascale Steven Zivanic Peter Urban #theCUBEresearch SiliconANGLE & theCUBE

31

5 Comments

✨Creating Custom Prometheus Exporters for OpenShift Workloads✨ I recently explored the guide on creating custom Prometheus exporters for OpenShift workloads, and I found it extremely practical and easy to follow. The document clearly explains why and when a custom exporter is necessary, making it easier to relate the examples to real-world use cases. The step-by-step instructions for building a Python-based exporter, deploying it in OpenShift, and configuring Prometheus to scrape the metrics were detailed and actionable. I particularly appreciated the sections on ServiceMonitor setup, Grafana integration, and visualization, as they provided a complete observability workflow. The best practices section was valuable for ensuring security, performance, and scalability in production environments. Overall, the guide empowered me to create custom metrics for proprietary applications and business-specific scenarios, enhancing monitoring and decision-making. I highly recommend this approach for teams seeking deeper insights into application behavior and operational performance in OpenShift environments. 👉Want the Creating Custom Prometheus Exporters for OpenShift Workloads PDF? Just drop a “PDF” in the comments and I’ll send over the link!📎 #OpenShift #DevOps #DevSecOps #Kebernetes #RedHat #OpenShiftTips #Prometheus #Observability #PrometheusIntegration #Learning #HandsonLabs #KubernetesMonitoring #CustomMetrics 

31

94 Comments

Here is a how-to blog on using the Argo CD Agent 0.4.1 with the OpenShift GitOps 1.18 operator for OpenShift users that want to try it out. As a reminder the Argo CD Agent is Technical Preview and not intended for Production workloads. However for customers interested in using it in the future it is well worth checking out now in a lab environment and providing feedback. I'm really excited about this technology, the OpenShift GitOps Engineering team along with the broader Argo CD community have done a really great job with the underlying architecture. https://lnkd.in/gYw_jyic

52

🔥 Announcing Microsoft Sentinel Model Context Protocol (#MCP) server 👉 A fully managed cloud service built on an open standard that lets AI agents seamlessly access the rich security context in your Sentinel data lake. 💡 Blog: https://lnkd.in/dMsND_6b

185

5 Comments

"If we look at #OpenShift, it's actually a platform answer." - said IBM Chairman and CEO, Arvind Krishna during the Earnings Call this week in response to a question posed about Red Hat. Proud to share the excerpt below where Krishna shares his perspective on Red Hat from the call transcript https://lnkd.in/gGNAKFkY "You asked me about Red Hat. So Amit, if you'll allow me, I'll go a little bit broader than the virtualization question. First, Red Hat #Linux, the core original product of Red Hat, is growing in the high single digits because of extreme demand around people deploying Red Hat in order to be able to leverage AI as well, not just the standard server footprint, which normally I would tell you gives you 6%, 7%, but additional workloads is adding that couple of percent that you're asking about. If we look at #OpenShift, it's actually a platform answer. So people are looking for what platform do we use for containers and what platform do we use for virtualization. There were 3 or 4 answers in each category. If I look at 2 of the 3 competitors on the containerization side, they've taken themselves out of the market in practical terms. So if people are wanting a container platform that goes across public cloud and on-premise, I will tell you that we are the leading answer for that. And every analyst as an industry analyst will tell you that. If you look at #virtualization, there are a few answers. But then there is a set of clients who would prefer a common answer across containerization and virtualization. And then by default, we tend to win those. That has accelerated the OpenShift piece. And last but not least, as we have brought HashiCorp in and the numbers show through in the HashiCorp product family, but the combination with #Ansible is very powerful and is going to boost Ansible as well going forward. So I hope this gives you a bit of color both on the macro and on Red Hat." Michael Barrett :: Ashesh Badani

89

2 Comments