The options announced by the third cloud provider target customers who need to maintain some specific workload in self-hosted or edge environments.
With its Distributed Cloud offering, Google Cloud has expanded its reach to enterprise data centers and the edge. The provider primarily targets companies with specific requirements for data sovereignty, latency or local data processing. Both the Google Distributed Cloud Edge and Google Distributed Cloud Hosted options announced this week are based on GCP’s Anthos product, which allows customers to deploy and manage Kubernetes workloads in diverse environments. “Google Distributed Cloud is a portfolio of fully managed software and hardware that expands our capabilities to data center customers,” said Sachin Gupta, vice president and general manager, at a press conference last week. of the open Google Cloud infrastructure.
The hardware will be provided by partners such as Cisco, Dell, HPE and NetApp. As for the software, it is based on the open-source container orchestration tool Kubernetes, originally developed by Google in 2014. The combined offer allows Google Cloud to better compete against rivals AWS and Microsoft. Azure, its customers with more choice in how and where they operate and manage enterprise workloads. “This initiative represents a strong alternative to Microsoft’s Azure Arc and AWS Outposts proposals,” said CCS Insight analyst Nick McQuire.
Google Distributed Cloud Edge
Called Distributed Cloud Edge, the first option allows customers to run applications in more than 140 Google edge global locations, in partner carrier environments, in their own data centers or colocation facilities, and in remote sites, such as the factory, while using multiple Google. data processing and analytics capabilities closer to where that data is generated and used. This solution is intended for applications that require low latency on the edge, with local data processing requirements that cannot be met in a public cloud, or for emerging 5G or LTE private workloads.
Expanding Google’s capabilities to the end is part of the current trend of cloud providers wanting to allow customers to connect more diverse devices – from factory sensors to industrial sensor devices – to a central hub where workloads can be cohesively managed. “The public cloud edge has a unique advantage: it expands the underlying cloud architecture and services, especially for customers who have invested heavily in a public cloud or otherwise,” Isaac wrote. Sacolick, President and CIO of StarCIO. “Are architects and developers looking to take advantage of AWS, Azure, or Google Cloud services deployed on the side? That’s the bet public clouds are making, which also has plans to expand to 5G-enabled mobile apps that require low-latency data and endpoint-level machine learning processing. telecom operators, “he added.
In this area, Google is a bit behind its rivals. But recently, the supplier has ensured to close this gap, particularly by launching Anthos for telecoms, by partnering with AT&T, Nokia and Ericsson on 5G connectivity, and by building edge its approach to the Google Mobile Edge Cloud cloud mode. Distributed Cloud Edge is available for preview now.
Google Distributed Cloud Hosted
The second option called Distributed Cloud Hosted is primarily focused on businesses that need to operate on a hybrid model and have strict local requirements or particularly sensitive workloads. According to Google, this solution will modernize on-premises deployments by accessing GCP services through a set of consistent APIs, but without a direct connection to Google Cloud. “We offer customers this hosted option to run in their own data centers to meet their low latency, resiliency, and privacy requirements,” said Thomas Kurian, CEO of Google Cloud.
Regarding the operation of Google Kubernetes Engine (GKE) on-premises by Anthos, Sachin Gupta explained that the difference is only in the control plane. When the clusters are “in-place or hosted, it is exactly the same as Kubernete but with a local control plane [fourni par Anthos]which means that Distributed Cloud relies on a full local control, “said Google Cloud’s vice president and general manager of open infrastructure. Future updates will be published by Google Cloud via dedicated secure repository, where the customer can review and validate them themselves before they airlift to apply in their local environment.
These options are in line with data sovereignty commitments made by GCP last year. “At Google Cloud, we take these issues very seriously, which are often addressed under the umbrella term of digital sovereignty. We work diligently in three areas: Data Sovereignty, Operations Sovereignty, and Software Sovereignty, to help address the issue of digital sovereignty in the cloud context. And we continue to listen to customers and make policy and incorporate their feedback in the best way going forward, “wrote Google Cloud CEO Thomas Kurian at the time. Distributed Cloud Hosted will not be available for preview until first half of 2022.