Google has had a busy week at its Cloud Next event in San Francisco, announcing new features and services across its portfolio.
Among the multiple new services announced at Google Next was an integration with GitHub and the Cloud Build CI/CD service. New services to help advance machine learning and analytics were also announced as well as the vision for an integrated set of capabilities that Google has dubbed the Cloud Services Platform.
"Talking to CIOs, they tell me they now realize they're going to be shutting down their data centers, and the other thing they say is looking at their workloads, there's just a tiny tiny fraction of them that are in the cloud," Diane Greene, CEO of Google cloud, said during the opening keynote. "So no wonder Google is just seeing amazing growth."
Cloud Services Platform
As part of the Cloud Services Platform, Google announced the 1.0 release of the Istio Service mesh open source project. Istio disaggregates microservices networking enabling services to be connected in a mesh.
While cloud services have largely to date been made up of virtual servers, there has been a trend toward serverless, event-based programming models as well. To that end, Google along with IBM announced the Knative serverless framework, which is a Kubernetes-based platform to build, deploy, and manage serverless workloads.
Google's Kubernetes Engine (GKE) is now also coming down from the cloud, with a new on-premises edition that enables organizations to run the same Kubernetes engine used in Google's cloud, on their own server infrastructure.
"Now, with GKE On-Prem, you can begin to modernize existing applications on-premise, without necessarily moving to the cloud," Urs Hölzle, Senior Vice President, Technical Infrastructure at Google, wrote in a blog post. "You gain control of your journey to the cloud at your own pace."
Control is also something that the new GKE Policy Management capability provides, enabling administrators to set polices across distributed clusters of Kubernetes deployments.
The core promise of the cloud has always been that it enables a consumption-based OpEx model for compute power. That model has long been reliant on different instance types that include certain amounts of CPU and memory. Now Google is changing things up with its resource-based pricing approach.
"Resource-based pricing considers usage at a granular level," Paul Nash, Group Product Manager, Google Compute Engine, wrote in a blog post. "Instead of evaluating your usage based on which machine types you use, it evaluates how many resources you consume over a given time period."
Sean Michael Kerner is a senior editor at ServerWatch and InternetNews.com. Follow him on Twitter @TechJournalist.