Google Cloud Anthos is a hybrid, cloud-agnostic container environment. Google Cloud Anthos is a software product that enables enterprises to use container clusters instead of cloud virtual machines (VMs) to bridge gaps between legacy software and cloud hardware. Google Cloud Anthos, sometimes shortened to just Anthos, was initially launched under the name of Google Cloud Services Platform and was later rebranded into Google Cloud Anthos in 2019.
The multi-cloud platform gives customers the option to use cloud technology such as containers and Kubernetes clusters on existing internal hardware -- since not all enterprises may want to abandon their existing infrastructure. Additionally, some enterprises may not be comfortable running all of their applications on shared services and may prefer hosting mission-critical application in their own data center. Typically a public cloud provider will offer a hybrid cloud that extends their services on premises; however, adapting those services to legacy systems which were originally designed to work in specific configurations might end up being extremely complicated, considering some applications may be specifically paired to certain on-premises servers. This problem can be solved through the process of virtualization, but this would sacrifice any of the benefits of hosting applications on a cloud-based infrastructure, such as scalability, efficiency and adaptability.
Giving users the option to use cloud technology on existing internal hardware is another way around this issue, which Google Cloud Anthos attempts to solve 'without sacrificing the benefits of cloud services. Offering a consistent design and set of services for both on-premises and in-cloud deployments, Anthos should give an organization freedom in choosing where to deploy applications, as well as migrating workloads between environments.
Google Cloud Anthos is aimed towards enterprise use, intended for use in managing and coordinating an enterprise's on-premises and cloud workloads. Organizations that want to pursue cloud-native applications with containers and microservices will find a lot of use for Google Cloud Anthos.
Having initially developed the open source container management system, Kubernetes, and building Google Cloud Anthos on top of a strong container-as-a-service platform should lead to a smooth migration path for legacy applications.
Google Cloud Anthos features
To create Anthos, Google partnered with hardware providers, such as Cisco, Dell EMC, Hewlett Packard Enterprise, NetApp and Robin.io to deliver the Anthos systems prepackaged in the software. However, Google Cloud Anthos can only run on servers capable of hosting Kubernetes clusters with the Google Kubernetes Engine.
Google Cloud Anthos aims to containerize legacy applications by including migration software that transforms VM images into containers before being deployed onto Anthos. The service is rounded out by a host of monitoring, security, authorization and application management tools. Some features in Google Cloud Anthos include monitoring, service management, a networking environment, Istio Service Mesh, centralized configuration management and consolidated logging. Google Cloud Anthos will also allow customers to manage workloads that run on third-party cloud services such as on AWS and Azure. This gives more freedom for users to run, deploy and manage cloud applications while positioning Google as a more neutral player compared to other cloud providers -- all while emphasizing workload portability. Potential customers are not therefore forced into only using Google's infrastructure. Administrators and developers are also not forced to learn how to work in different environments with different APIs. In addition, Google Cloud Anthos is based on Google Kubernetes Engine (GKE), Google's managed Kubernetes service, meaning users will get any feature updates and security patches.
Components of Google Cloud Anthos
Google Cloud Anthos is created using multiple systems. Anthos' core, however, is a container cluster that is managed by Google Kubernetes Engine. To accommodate hybrid environments, Google Cloud Anthos supports the Google Kubernetes Engine managed container service as well as a GKE On-Premises environment, which bundles the same set of management and security features.
Other components of Google Cloud Anthos include:
- Anthos Config Management: This provides the tools required to set up and administer multiple Kubernetes clusters. Config Management also allows users to maintain a consistent set of network and security policies across multiple environments. It will manage clusters via configurations that are stored in repositories such as GitHub or Google Cloud Source Repositories.
- Traffic Director: This is the traffic control plane for a service mesh that notably adds multi-region load balancing, health checking and demand-based autoscaling. It uses standard APIs that enable it to work with sidecar proxies, like Envoy.
- Stackdriver: This is Google Cloud Platform's (GCP) monitoring, management and debugging service. Stackdriver provides a central repository for network, application and infrastructure telemetry. Google offers two options for support, which are Legacy Stackdriver and Stackdriver Kubernetes Engine Monitoring.
- Migrate for Anthos: This automates the migration of legacy virtualized workloads to GKE containers. Migrate for Anthos will also convert workloads from VMware vSphere, Amazon EC2 and Microsoft Azure VMs.
- Cloud Run:Â This is a serverless platform that can run event-driven workloads as well as deploy containers to a GKE cluster. Cloud Run will enable apps to use container-run functions without configuring servers. Additionally, Cloud Run can also automatically size compute resources according to workload demands.
- GCP Marketplace:Â This allows Anthos users to access prebuilt Kubernetes development stacks and applications on a GCP Marketplace. The marketplace can also automate the container development processes using CodeBuild.
- Apigee:Â This is GCP's API management service. Apigee aids in simplifying the process of exposing programming interfaces for services that run on GKE.
- Google Kubernetes Engine: This is a management and orchestration system for Docker containers and container clusters that run from within Google's public cloud services. Google Kubernetes Engine is based on Kubernetes. GKE can be used to create container clusters, container pods, replication controllers and load balancers, as well as resize container clusters and application controllers. GKE can also be used to debug, update and upgrade container clusters. Users can also use GKE to migrate existing VMs into containers.
- GKE On-prem:Â This is also known as Google Kubernetes Engine-Premises, which delivers Kubernetes-based software to an organization's hardware on-site. The platform is then managed by Google.
- Istio on GKE: This is a service mesh that securely connects clients to containerized applications and services. Istio on GKE will also manage traffic flows between microservices, as well as enforce security and usage policies. Istio specifically enables developers to connect, control, observe, secure and run distributed microservice architectures, regardless of platform or vendor. Istio can manage service interactions across containers and virtual machine workloads.
- Velostrata: This is a cloud migration technology that Google acquired in 2018 that is used to stream on-premises physical and virtual machines, creating replicas in GCE instances. This is used to convert VMs to Kubernetes Pods. In other words, Velostrata is a physical-to-Kubernetes migration tool. This is part of Anthos Migrate.
- GCP Cloud Interconnect:Â This component strives to deliver high-speed connectivity between enterprise data centers and the cloud infrastructures. GCP Cloud Interconnect can deliver high-speed data up to about 100 Gps. Users are also able to use other networks from separate vendors that can extend their data centers.
Anthos Platform
The Anthos Platform is essentially another way to determine which services in Google Cloud Anthos can be used in what environments. There are three segments to the Anthos Platform; these include Application Operation, Network and Security Operation and Platform Operation -- all of which are available on premises or in the public cloud.
Application operation is generally used in hybrid environments and includes:
- continuous integration and continuous delivery tools
- serverless tools
- Cloud Run for Anthos
- Apigee
Network and Security Operation is typically used to implement policy, security controls and compliance for on-premises and in public clouds. Network and Security Operation includes:
- Anthos Config Management for policy and configuration management;
- Istio for policy and configuration management;
- Istio for zero-trust network security;
- service discovery and management -- using Anthos Service Mesh;
- a secure software supply chain; and
- multi-tenancy -- using GKE Sandbox.
Platform Operation is typically used to centralize cluster management. It is also used to standardize with Kubernetes and provide a consistent service across multiple environments. Platform Operation includes:
- Google Kubernetes Engine for managing Kubernetes and centralizing management;
- tools such as a traffic director for networks;
- GKE storage options;
- on-premises storage options;
- Migrate for Anthos -- to move workloads directly into containers on Google Kubernetes Engine; and
- monitoring using Stackdriver Kubernetes Engine Monitoring.
Future
The future of Google Cloud Anthos looks like it may survive over other cloud computing platforms for its flexibility in functioning -- giving customers a choice in where to deploy, manage and run their applications, on either their own hardware or in a cloud service. This flexibility can fee an organization up so they are not locked into a single cloud provider and can shift workloads to different cloud services when needed.
One competitor to Google Cloud Anthos is Microsoft Azure, which, in itself, has staying power. Azure is Microsoft's public cloud computing platform. Azure provides cloud services including those for compute, analytics, storage and networking. Users can also choose from the services to run existing or develop new applications in the cloud. Both services have their benefits, so potential customers should look into both before picking either service. Still, Google Cloud Anthos' reliability and flexibility makes its mark in having potential staying power.
Pricing
The pricing of Anthos is based on a monthly subscription that requires a minimum one-year commitment. However, its prices make it clear that it is aimed at larger enterprises. The price listed for Anthos is $10,000 per month -- per 100 virtual CPUs (vCPU). This cost expense only pays for the Anthos managed software layer, not including the underlying infrastructure on Google's cloud. Anthos is also sold in 100-vCPU blocks.
Prices for Google's cloud infrastructure may also vary across instance types, and, more confusingly, cost can vary based on geographic region. As an example, in 2019, the monthly cost for a low-end, n1-standard-1 instance including one vCPU and 3.7 GB of memory is $24.27 per month in Google's Iowa region, but $33.98 in its Hong Kong region.
Additionally, support is not included in the monthly cost of Anthos -- even though Google requires it. Google also recommends an enterprise-level support contract, which costs either $15,000 or a percentage of a customer's total spending.
Â
Â