devopsy.kinsta.cloud

Tag: Google Anthos

  • HPE Extends Hybrid Cloud Computing Strategy

    HPE Extends Hybrid Cloud Computing Strategy

    Hewlett-Packard Enterprise (HPE) today advanced its hybrid cloud computing strategy by announcing at its HPE Discover 2019 conference an alliance with Google to embrace Kubernetes in hybrid cloud computing environments and then extending the reach of its software-defined infrastructure for building private clouds to legacy ProLiant servers.

    Lauren Whitehouse, director of marketing for the software-defined and cloud group at HPE, said the HPE Composable Cloud now extends beyond new ProLiant servers to include everything from legacy hardware to cloud computing platforms running virtual machines or as bare metal servers optimized for Kubernetes.

    To achieve that latter goal, HPE will make it possible to deploy Google Anthos, a previously announced framework for delivering cloud-native computing services across public clouds and on-premises IT environments, on HPE-managed platforms.

    HPE is giving customers the option of deploying configuring local servers with virtual machines with either a VMware vSAN storage software or the HPE SimpliVity hyperconverged infrastructure platform. Additionally, customers can choose to deploy Kubernetes on those platforms or run Kubernetes on a bare-metal server. On top of those stacks, HPE also embeds the HPE OneView IT infrastructure management and HPE Composable Fabric networking solution to create what the company calls a composable rack environment.

    Finally, HPE announced it will extend the reach of the HPE InfoSight artificial intelligence engine, which it gained for storage environments when it acquired Nimble Storage, to the rest of its data center portfolio.

    Whitehouse said that, by extending the reach of its composable computing strategy to include legacy servers, HPE seeks to make it easier to modernize on-premises IT environments within the context of a larger hybrid cloud computing strategy.

    One of the primary reasons so many workloads are being deployed in the cloud is because of the inherent flexibility of the IT environment. HPE is investing in higher levels of infrastructure abstractions to enable an equivalent amount of agility to be injected into on-premises IT environments. Those efforts will then make it easier for organizations to extend best DevOps practices across a hybrid cloud computing environment, said Whitehouse. The company is now removing the objection by organizations to have to implement a forklift upgrade across their entire on-premises IT environment to achieve that goal, she added.

    It’s not immediately clear who will manage what in this brave new world of hybrid cloud computing. Application developers have been exercising a lot more influence over platforms, while advances in automation appear to be heralding a consolidation of management functions within the local data center. As site reliability engineers, or so-called “super administrators” capable of managing compute, storage and networking infrastructure continue to emerge, the silos around job functions in the enterprise were built are breaking down. In fact, the more organizations embrace DevOps best practices, the faster consolidation of job functions is occurring.

    In the meantime, whoever provides the infrastructure that remains in an on-premises IT environment had better be able to enable local IT teams to be just as flexible as any cloud service provider an application developer is likely to encounter.

    — Mike Vizard

  • Google Revamps Hybrid Cloud Strategy

    Google Revamps Hybrid Cloud Strategy

    At the Google Cloud Next 2019 conference, Google announced it will make available cloud services based on Kubernetes under the brand name Anthos on its own cloud and on-premises platforms, as well as on third-party cloud services. such as Amazon Web Services (AWS) and Microsoft Azure. Google also announced a tool dubbed Anthos Migrate that enables any virtual machine to be encapsulated in a container that then can be migrated on to Anthos Migrate. More than 30 vendor partners, including Cisco Systems, Hewlett-Packard Enterprise (HPE), Lenovo, Dell EMC, VMware and Intel, announced their support for the latest expansion of the Google hybrid cloud strategy.

    Google also is now making available in beta a runtime for the Apigee API management platform that can be used anywhere. Dubbed Apigee hybrid, Google acquired the API management platform in 2016.

    In addition, Google is making available a managed set of database services in collaboration with Confluent, MongoDB, Elastic, Neo4j, Redis Labs, InfluxData and DataStax.

    The company also announced Cloud Run, a managed serverless computing offering based on the Knative middleware Google developed to make it possible for applications running on Kubernetes clusters to dynamically invoke a serverless computing framework. Google also announced it is updating its App Engine serverless computing framework to add support for Node.js 10, Go 1.11 and PHP 7.2 runtime in general availability and Ruby 2.5 and Java 11 runtimes in alpha.

    Finally, Google updated both its functions programming framework and virtual private connection (VPC) software used to create virtual private networks, while also revealing it has added regions in Salt Lake City and Seoul, South Korea. In all, there are now 22 GCP regions.

    Like other cloud service providers, Google continues to invest heavily in managed services through which it essentially functions as the IT operations team for application developers. In the case of the databases Google is making available as a managed service, those offerings are distinguished by the fact that they are being delivered in collaboration with the companies that oversee the open source projects that maintain and update those databases.

    DevOps teams will, of course, need to determine to what degree they want to deploy and manage open source databases in the cloud versus relying on Google and its partners. Mavinder Singh, head of infrastructure partnerships for Google Cloud, predicted the long-term trend will be to rely more on databases delivered as a service to allow organizations to focus more resources on developing applications.

    The challenge DevOps teams increasingly face is that next phase of cloud computing is considerably more complex than the previous phase. In the first phase of cloud computing most organizations replaced an on-premises IT environment based on virtual machines from VMware with virtual machines made available of cloud infrastructure. The second phase is more complex in that monolithic applications are now being eclipsed by so-called cloud-native applications based on instances of Kubernetes clusters running on top of virtual machines that are augmented by serverless computing frameworks. Most IT organizations don’t yet have the skills required to deploy, manage and secure those cloud services on their own. In time, however, as it becomes easier to automate the provisioning of the platforms required to run cloud-native applications, many IT organizations may prefer to rely on their own internal DevOps team versus contracting for a managed service, especially if those managed services don’t conform easily to the DevOps processes those organizations have adopted.

    It may take a while for organizations to determine which path forward ultimately suits them best. But the one thing that is clear is that cloud service providers are investing heavily in providing managed services direct to developers to add value on top of what have rapidly become commodity IT infrastructure services.

    — Mike Vizard