Red Hat OpenShift Red Hat OpenShift Data Foundation | Red Hat OpenShift, 28 Feb 2022

Previous Meeting

⏯

youtube image

►

From YouTube: Red Hat OpenShift Data Foundation overview

Description

Red Hat OpenShift enables organizations with the ability to build, deploy and run applications on any infrastructure or cloud. Complex, mission-critical applications are now extended across the hybrid cloud and need integrated persistent storage. Red Hat OpenShift Data Foundation provides these data services to support stateful applications and data-intensive workloads. Find out more about what OpenShift Data Foundation provides, why it is important to applications running in Red Hat OpenShift, and how organizations around the world are benefitting from this solution.

Learn more: http://red.ht/RHODF

A

Thanks for joining me for this quick overview of red hat openshift data foundation, now in this session we will discuss what openshift data foundation is, why it benefits applications running in red hat, open shift and how organizations are using it to deliver cloud native services.

A

So, let's start with what makes up openshift data foundation. Openshift data foundation is made up primarily of three technologies. The first is rook, which works with kubernetes to provide the orchestration layer to openshift. This is what provides the simplified day, one and day two operations for the platform. It's integrated with the openshift console and it delivers automated administration and management.

A

Next openshift data foundation has at its core software defined storage technology. Now this is what provides the persistence for any stateful applications that you're developing in openshift and importantly, it provides file block and object storage classes.

A

So, no matter whether you need a shared file system for your applications or you're running structured data workloads like databases that need a high performance block based data backend or maybe you need massively scalable object, storage for cloud applications and for emerging technologies like artificial intelligence and machine learning.

A

Openshift data foundation provides all of these protocols natively within one data service platform and then to support a consistent experience across your hybrid and multi-cloud deployments. Openshift data foundation includes a technology called nuba.

A

Now this delivers a multi-cloud object, gateway that supports data mirroring spreading tiering object, bucket claims and data federation with access to name space buckets, no matter where the client resides on premise or in a private or public cloud.

A

Okay. So where does openshift data foundation sit well? Red hat. Openshift is the core platform that delivers services to cloud native applications, and these are platform, services, application, services, developer and data services, and these data services focus on delivering things like data efficiency, resiliency security, governance and data discovery for those applications that are running in red hat openshift.

A

So you might think of things like compression and performance data protection, disaster, recovery policies, encryption data, immutability and compliance tagging, searching federation just a few examples, but to provide a robust, production-ready environment. All of these data services must be covered and integrated with the platform, and that's what openshift data foundation delivers and organizations are adopting red hat openshift for more complex and mission-critical workloads.

A

They need a solution that completes the platform tools, such as multi-cluster management, cluster security, a global registry and cluster data management can't be seen as an afterthought to your application, modernization strategy. That's why we bundled all of these essential components in one product, and that is red, hat openshift platform plus, which includes openshift data foundation.

A

Okay, so now that we have a good understanding of what openshift data foundation is and what it delivers, let's take a look at why it's so important for your applications running in red hat openshift.

A

So, first of all, it provides simplified access to data, and this is not only from an operations perspective by simplifying the deployment, automation and management of the data platform, but it puts all of your data resources in the hands of the user, whether that's a developer, an analyst data scientist or even an automated process, enabled by eventing now. Another huge benefit is that it delivers a consistent experience.

A

Regardless of the underlying infrastructure, so it supports the data workflow from the edge to core and out to the cloud all with the same openshift management console and all with the same user experience, regardless of where your data resides and then due to the robust architecture of the solution, openshift data foundation delivers massive scalability and, more importantly, it has proven deterministic performance at scale, and this has been demonstrated time and time again in actual deployments by partners and by independent analysts.

A

Openshift data foundation makes data accessible to applications running across the hybrid cloud, driving new insights, unlocking the potential of emerging technologies and modernizing the way that organizations work.

A

In the end, we enable innovation without limitations, so just to show you how easy it is to install openshift data foundation from your openshift console, you just have to go into operator hub and get the operator now, a quick way to find it is to just type in odf for openshift data foundation, and then here you just click on install and you're on your way. Now, once the operator is installed, you start off by creating the storage system.

A

Here you choose the storage class and how large you need your cluster to be, and you provision the nodes you want to be part of your storage cluster and then click next.

A

Okay, here you have some options for encryption and network that you can change if you'd like and now just verify, choices and click on create storage system. Okay, so after your cluster has been created, you can go into your openshift console and then into the openshift data foundation. Dashboard and from here you can manage persistent volumes and object buckets.

A

You can see here, block and file block, pools and object, storage resources that you've built for applications across the hybrid and multi-cloud, and whether you're deploying on bare metal virtual machines across a private cloud or on one of the public cloud services.

A

You're assured of a consistent user experience with dynamic and static provisioning models, and not only are you protected from failures within an availability zone, but you also have persistence across failure zones and no matter what your workload is. Openshift data foundation supports all data storage protocols, all within one solution, integrated with and managed by red hat openshift.

A

Likewise, you can read about how openshift data foundation and ceph perform with different applications and workloads. You can even see how we perform against competitive products in real life workload benchmarks from independent, analysts, okay, so that brings us to the how portion of our presentation, how are organizations across the globe using openshift data foundation?

A

Well, we look closely at data engineering patterns in respects to workloads and we've identified three data trends as data at rest, and this is stored. Data in repositories, databases or on file systems object data data in motion.

A

Now this is data, that's moving or transitioning across applications or infrastructure, triggering events or being transformed for consumption and then data in action where data is actually being worked on or processed for things like business, insights, model, training alerts or notifications, for example, and so it's important to understand how our platform works with data in these various states from a resiliency perspective, scalability and, of course, performance, but it isn't sufficient to just run a workload and check a box that says it's validated.

A

We've undertaken projects that test real-world scenarios, even attempt to break things and create what-if scenarios. So that we learn best practices as well as where openshift data foundation is best suited and where we need to focus our efforts for our product roadmap.

A

There are several architecture detail papers that focus on different workloads for data rest data in motion data and action and openshift data foundation also provides a robust data layer for all of your infrastructure services in open shift so consider things like monitoring and metrics logging and registry infrastructure services such as quay that require multi-protocol storage classes.

A

Openshift data foundation delivers the data services layer for your openshift cluster, in addition to all of your application workloads and we build services with openshift to enable a robust data, protection and disaster recovery strategy.

A

We've validated several of our data protection ecosystem partners, products to give customers a breadth of choice, including integration, with existing data center, backup and recovery solutions and we're working with customers on metro and regional disaster recovery solutions that drive down recovery, point objectives and recovery time objectives to close to zero, enabling the most robust uptime service levels for your most critical applications.

A

So hopefully, you've gotten a good sense of what openshift data foundation is why a data services approach is important for your data in kubernetes and how you can use it for marriott workloads.

A

Now, I'd like to just take a few moments to introduce you to a project that we're collectively working on at red hat, to help organizations accelerate innovation for emerging industry, use cases using many of the tools that are available in red hat openshift amq openshift data foundation, as well as leveraging our robust partner ecosystem.

A

So data engineers they get quite a bit of product documentation, but what they don't typically get is a lot of documentation or sample code on how to compose end-to-end data pipeline solutions from multiple products.

A

Now, to that end, we've been developing several data engineering, validated patterns based on real-world scenarios, so data engineers can start working with sample solutions using modules that can be used and reused to develop new solutions to specific use cases, and we have a few of these out there today, ranging from a manufacturing use case for preventative maintenance using industrial sensors, a horizontal use case for managing pipelines across clouds and several others that we're developing. But I thought it would be interesting to give a quick overview of a healthcare use case that we've been developing.

A

So essentially, what we're looking at here is automating a data pipeline from an edge device, in this case an x-ray image in a clinical setting and then using artificial intelligence to assist the diagnosis in this case pneumonia and then using red hat solutions to stream and trigger events that alert medical staff and even enable machine learning model retraining. So the system improves with time in this sample use case.

A

Our goal is to improve the time and accuracy of patient diagnoses so that the practitioner can focus and respond quicker to cases and to give a little bit more depth into the architecture of the solution. You can see here that the x-ray images are ingested into an object bucket and openshift data foundation.

A

The rados gateway within ceph then spawns off bucket notifications to an endpoint, in this case a kafka endpoint in red hat amq now from there k-native, which is the technology behind openshift serverless spawns off a number of activities, including an inference function that uses artificial intelligence to detect probability that the image depicts a pneumonia case.

A

If that probability is high, it immediately notifies the medical staff so that they can respond more quickly to high-risk patients and it also anonymizes the image, removing any personal information and sends that image back to the core data science team to help retrain that ai model, and so with that, I'm going to leave you with a few short urls. So you can follow for more information to get the sample code for the validated patterns or even get an evaluation copy of red hat openshift data foundation.

A

And, of course, if you have any questions or you'd like to chat about any projects that we might be able to help you with, please don't hesitate to reach out we'd love to hear from you thanks. So much for your time today,.

A