This is a crucial time to be developing new cloud technologies for solving problems. Cloud is now being used for navigating ships carrying cargo, connecting online and offline retail markets, machine learning in fintech for fraud detection and many more.
The computational demand changes for applications and users. There needs to be a flexibility between different demands and still be able to deliver results in real time.
The key functionalities of AI include vision, speech, prediction or forecasting. Many organisations now deploy vision and speech systems like a routine. For instance, recommendation engines requires both knowledge and tools. The knowledge of building a pipeline and the tools to handle data ingestion, storage, security, access and deployment.
To provide this flexibility at large scale, Google recently launched Anthos, a cloud product at its recently concluded next event.
Anthos is based on Kubernetes, which is an open source technology. It is capable of running on hybrid cloud as well as on-premise.
Anthos lets the user run their application without any modifications and also manage workloads running on third-party clouds like AWS and Azure. This level of freedom enables developers to deploy applications on cloud without the need to learn new APIs.
Many global organisations require hybrid functionality. They need both on-prem and cloud options to manage their services. So far, companies like HSBC, Siemens have been benefiting from this platform while other partners have promised to imbibe Anthos into their infrastructure.
What Does Anthos Get Right
To reduce complexity Anthos enables decoupling of infrastructure and other services. This reduces cost and increases developer velocity. This is achieved with the help of Kubernetes and containers.
Containers provide isolation among different applications in the same machine.An application is packed independently from the infrastructure along with the libraries and other dependencies.
As the complexity of applications grows, pods of Kubernetes come into play. Pods form a cohesive unit of containers consisting of multiple processes. The decoupling becomes easy as these pods contain the IP addresses of each of the constituents in the cluster.
For configuring these applications, Google uses restricted language YAML and is automated using controllers. This eliminates the failures and complexities that arise from using other languages like Python.
Along with scalability and simplicity, Anthos also has to ensure the security of the transaction. And, Google has pioneered the zero-trust access security model to maintain consistent user experience. Along with this, the proxy-gateway approach also helps to ensure the service providers to efficiently deploy APIs on cloud in the backend.
Key Takeaways From Decoupling
The team behind Anthos at Google Cloud, lists the following properties that enable play on go kind of feature for Anthos:
- Basic services in Kubernetes that decouple services from pods, allowing each to evolve independently.
- Istio proxies to provide capabilities, including security and telemetry, across all services in a uniform way that is decoupled from each services’ source code.
- Proxy-enabled policy deployment and enforcement decouples the management of policies from specific services so that policies can evolve without service disruption and security posture can be improved without infrastructure changes (via policy as code)
- The use of services as the unit of deployment that often correspond to specific teams, allowing teams to operate more independently. This may result in hundreds of services for a large-scale application, the Kubernetes and Istio service infrastructure makes it simple to manage.
Anthos can be used as a gateway to have a more secure future for any business as it enables building of applications on top of its platform that allows you to run tasks without the need to stack old sources.
This increases the speed at which a business is conducted and maximises resource utilisation.
Know more about Kubernetes’ Pods here
Register for our upcoming events:
- Meetup: NVIDIA RAPIDS GPU-Accelerated Data Analytics & Machine Learning Workshop, 18th Oct, Bangalore
- Join the Grand Finale of Intel Python HackFury2: 21st Oct, Bangalore
- Machine Learning Developers Summit 2020: 22-23rd Jan, Bangalore | 30-31st Jan, Hyderabad