- Integrated Systems
- About Us
- Integrated Systems
- About Us
Re: Introducing HPE Ezmeral Container Platform 5.1
Over the last few months, HPE has made considerable progress towards our goal of providing everything-as-a-service for our customers, from edge-to-cloud. We made two recent announcements that represent key milestones towards that goal. In June, we launched our new software brand to unify our software portfolio: HPE Ezmeral. At the same time, we announced new HPE GreenLake cloud services – for Containers and MLOps – delivered and fully managed as-a-service by HPE, powered by our HPE Ezmeral Container Platform software.
With these announcements, HPE has outlined a robust software and cloud services strategy to help our customers accelerate their data-driven digital transformation initiatives. The HPE Ezmeral Container Platform is one of the cornerstones of this strategy. With the HPE Ezmeral Container Platform, we provide our customers with:
- A unified multi-cluster control plane: To deploy, run, and manage multiple Kubernetes clusters in hybrid cloud and multi-cloud environments.
- A platform to modernize both cloud-native and non-cloud-native monolithic apps: With differentiated capabilities to containerize monolithic and stateful applications, without the need to refactor code.
- Ready-to-run workload-optimized solutions: To build containerized applications for a wide range of use cases, with their choice of open-source or ISV tools, and deploy them anywhere – on-premises, in any public cloud, or at the edge.
To that end, I’m proud to introduce a new release of our container platform with exciting new capabilities in each of these areas.
Unified control plane
External Kubernetes cluster management: Hybrid cloud is quickly becoming the norm for enterprise IT; our customers are looking for solutions that will allow them to build anywhere and deploy everywhere. However, it is extremely challenging to monitor, manage, and optimize applications across multiple cloud environments. Users have to log into different management consoles to manage Kubernetes clusters across their hybrid IT landscape. Now our container platform offers the flexibility to deploy and manage any Kubernetes cluster from a single pane of glass – either for on-premises Kubernetes deployments and/or on public cloud providers’ Kubernetes services.
With the 5.1 release, HPE Ezmeral Container Platform provides a unified control plane to deploy applications to Amazon's EKS clusters or to any Cloud Native Computing Foundation® (CNCF®) certified Kubernetes cluster – with extension to Google Kubernetes Engine (GKE) and Azure Kubernetes Service (AKS) clusters.
Platform administrators get a single pane of glass to dramatically simplify the management – resource allocation, user authentication, workload deployment, and monitoring – for any Kubernetes cluster. And end users can use the built-in App Store to deploy their apps on any Kubernetes cluster – on-premises or in the public cloud – with a few mouse clicks.
Massive Scalability: We have built our platform to be enterprise-grade and massively scalable, and we’ve proven this out in the real world with some of the largest and most demanding enterprises – for both dev / test and production environments. For example, one of our customers – a large financial institution — has deployed a single cluster of more than 1000+ nodes for a log analytics application. With this release, we’re continuing to make our platform even more robust for large-scale deployments – and we’re committed to providing best-in-class scalability in our upcoming releases as more and more customers expand their container deployments at scale.
Trusted private registry for applications: This is good news for teams that are looking to securely store application images: we are integrating the open source Harbor registry into our platform. Users no longer have to live with the security vulnerabilities of unknown application images. The HPE Ezmeral Container Platform scans images for security vulnerabilities with image signing – where certified, trusted images can be rapidly and securely deployed. And we have implemented role-based access controls (RBAC) so each tenant gets its own secure workspace within the registry.
Modernization for stateful applications
Istio service mesh integration:
Although cloud-native applications offer a wealth of benefits, they can be challenging for DevOps teams that are managing these applications in a hybrid cloud environment. Admins need to configure to manage the traffic between microservices, monitor performance, upgrade or patch each service, and so on. This can get very challenging when you have a large application estate to manage. This is where Istio comes in. Istio simplifies how you connect to, secure, control, and monitor microservices, and thereby eases the strain on DevOps teams.
We are integrating the open source Istio service mesh and providing it as an out-of-box capability on the HPE Ezmeral Container Platform, enabling a service mesh on each cluster while providing the flexibility to selectively deploy Istio on a per-tenant basis. With tenant isolation, we provide granular, tenant-level observability allowing each tenant to independently monitor applications running on Istio. Tenant and platform admins can enable Istio at the tenant or cluster level with a single mouse click, thereby allowing IT teams to deploy applications that are service mesh aware while also supporting applications that do not utilize a service mesh.
Global shared volumes: One of the essential requirements and challenges to deploying stateful applications on containers has been access to shared data, models, and other related artifacts across projects. Stateful applications that need to run in containers need persistent storage to store intermediate application data. Also, most stateful enterprise applications require access to enterprise data. The global shared volumes capability leverages HPE Ezmeral Container Platform’s DataTap capability to provide secure access to shared data across applications. Applications running in different tenants can share code, data, and other artifacts across tenants. HPE Ezmeral Container Platform includes pre-integrated HPE Ezmeral Data Fabric for persistent container storage (private volume) and the shared global storage (global volume). Tenant admins can set up global shared volumes and enable user access with just a few mouse clicks.
HPE Ezmeral Data Fabric on Kubernetes: With version 5.0, we incorporated a global, edge-to-core, high-performance, multi-protocol (NFS, POSIX, HDFS, S3, CSI) data persistence layer with the pre-integrated HPE Ezmeral Data Fabric. With the 5.1 release, we deliver the same capabilities as native Kubernetes storage. The HPE Ezmeral Data Fabric runs as a workload on a Kubernetes cluster. This makes it even easier to install, upgrade, patch, and scale data fabric clusters while simplifying the lifecycle management for these clusters.
Our vision for the HPE Ezmeral Container Platform is to provide workload-specific solutions for a variety of different enterprise use cases while providing a common platform to manage these different workloads. Currently, customers can license HPE Ezmeral ML Ops – a solution built on our container platform – to operationalize machine learning and deep learning models at scale and move from pilot to production with their AI initiatives. We plan to introduce additional workload-specific solutions (e.g. for Data Ops and other use cases) over time.
Kubernetes-based applications: With support for open-source Kubernetes, customers can deploy any Kubernetes-based application on the HPE Ezmeral Container Platform. However, it can be challenging to deploy and manage all of these different applications and use cases – with all the right tools, frameworks, and components. Our goal is to make this easier for our customers, by packaging the most common tools for specific use cases and applications. With the 5.1 release, we are integrating (out-of-the-box) and fully supporting the Spark and Kubeflow operators; we’ll also support a wide variety of Kubernetes-based machine learning tools such as Argo, Seldon, and many more. We’ve built this capability into our HPE Ezmeral ML Ops solution, to provide data science teams with the flexibility and choice to deploy and operationalize ML models using their tools of choice.
Supporting DevOps processes: As another example, we’re also enabling our customers to build Continuous Integration / Continuous Delivery (CI/CD) workflows and processes by providing them with source control and workflow automation tools such as Github (available in 5.0), Bitbucket Enterprise, and Jenkins (available in 5.1).
These are just a few examples of the functionality and innovations in this exciting new release of the HPE Ezmeral Container Platform – with much more to come.
I encourage you to learn more about the HPE Ezmeral Container Platform by checking out our on-demand sessions and demos at the HPE Discover Virtual Experience event, including:
Plus, we have additional sessions and live demos lined up this week on July 29 (registration is free here), including:
For more information visit the HPE Ezmeral Container Platform web page.
Hewlett Packard Enterprise
Matheen Raza is a Senior Product Marketing Manager at Hewlett Packard Enterprise (HPE) in the Enterprise Software Business Unit. A technology enthusiast, he is extremely passionate about cutting edge technology such as containers, AI, machine learning, and deep learning to improve business outcomes in enterprises.
- MTiempos on: HPE Ezmeral Container Platform is now HPE Ezmeral ...
- Alexander6 on: Three advantages of a data fabric for retailers
- on: Analytic model deployment too slow? Accelerate dat...
- Jeroen_Kleen on: Introducing HPE Ezmeral Container Platform 5.1
- LWhitehouse on: Catch the next wave of HPE Discover Virtual Experi...
- jnewtonhp on: Bringing Trusted Computing to the Cloud
- on: Leverage containers to maintain business continuit...
- on: How to accelerate model training and improve data ...
- vanphongpham1 on: More enterprises are using containers; here’s why.
- on: Machine Learning Operationalization in the Enterpr...