Multi-cloud redefined by Google

No alt text provided for this image

Ask anyone from the IT industry or anyone who observes the IT industry to name the top trends in the industry which they see, You will inevitably hear from everyone many different things on the theme of cloud computing. In particular, you can notice everyone will talk about multi-cloud computing.

Multi-cloud computing is simple that companies are choosing to use different cloud services such as Google’s GCP, Amazon’s AWS, Microsoft’s Azure, etc. This approach is chosen by the companies to avoid a service provider lock-in, this helps in redundancy and also deals with data-related issues where some data has to be stored within a given country for legal and another regulatory purpose.

In addition, Some of the cloud providers have expertise in specific areas and companies usually choose to run their workloads with these specific providers to have the advantage of their unique capabilities.

Even in those specialized situations, the datasets are totally different, application and workload that is run in those are different. The announcements that came from Google Cloud CEO Thomas Kurian made at Google Cloud Next virtual event, it is very clear that Google is planning to expand the definition of multi-cloud.The new offering that Google announced works across multiple datasets and workloads which are running on different cloud service providers. The new BigQuery Omni capability will let companies to leverage Google’s deep heritage in search capability and query capability to run the analytics power across the data which is stored in Google Cloud, AWS, and, later in 2021 Azure.

Let’s understand at a very high level what BigQuery Omni helps us with

BigQuery Omni primarily permits a macro-level grouping of multiple cloud platforms, workloads, and datasets beneath one level of control, all of this is done by leveraging the widely accepted strength that Google has on its competitors. This combines multi-cloud, data analytics architecture which is a very aggressive and smart move that is making Google to win, and the big consumers will look up to Google.

one of the key benefits of using BigQuery Omni will be that Google code runs natively on the different cloud platforms and it will access data from the local storage resources within those platforms, this will avoid the cost and time loss of processing the data transfer across platforms(data egress). The BigQuery Omni users can utilize the same BigQuery interface (running in GCP) to create the necessary SQL commands to query the databases and the requests are computed locally in each of the environments. The results which we get out of multiple sources will be then transferred back to one single UI for easier analysis or can also be stored within each platform to avoid cross-cloud data movement.

This is all potential is that since its origin over ten years agone as a Google internal tool, BigQuery, with its Dremel search engine, the computer element is separated from the storage element. This was not built keeping in mid the multi-cloud computing, but Google combined Google Anthos aboility to run or transfer the workload across the cloud service providers through an abstraction layer, this solution actually treats different cloud service provider as if they were different regions within a single platform.

Thus getting an extended multi-cloud version of a single cloud platform, with BigQuery Omni which seems to offer a lot of interesting benefits. For a user the meta platform concept providers a compatible way to access multiple data sets across different cloud providers with a single interface. Thus making cloud computing becomes more flexible. Google(or other cloud providers) lets the user leverage some of the unique IP which is given by other Cloud vendors like (AWS,Azure). Algorithms across multiple datasets stored in multiple locations can be used for ML or neural network training.

By using features given by both BigQuery and Anthos, google now has a new wonderful twist on the multi-cloud. with BigQuery Omni in place, we have the potential to open up a multip-cloud world which is totally unified macro-level, meta platform which could give way for a new powerful future with the help of Google. It remains to be seen however effective they persuade be within the universe, however conceptually it breaks some fascinating new ground.





Rajesh Angadi

Consultant - Technology driving digital innovation , Operational efficiency with Main Character Energy

3 年

Congratulations Dr.Muthukumaraswamy!

回复

要查看或添加评论,请登录

Dr.MuthuKumaraswamy B的更多文章

  • AI Agents in Enterprise Production: Strategies for Success

    AI Agents in Enterprise Production: Strategies for Success

    AI and ML-driven agents are revolutionizing enterprise environments, but deploying them effectively in production…

  • Inside the Mind of an AI Agent: How It Works and Why It Matters

    Inside the Mind of an AI Agent: How It Works and Why It Matters

    Autonomous agents operate much like humans, relying on fundamental components such as identity, memory, planning, and…

  • Faster way to learn programming languages

    Faster way to learn programming languages

    Getting better at something faster isn't about skipping topics and being half-baked. I'm not showing you any shortcuts.

    3 条评论
  • Explaining AI: What Do Business Leaders Need to Care?

    Explaining AI: What Do Business Leaders Need to Care?

    AI and the challenge of model explainability. The use of artificial intelligence (AI) has become more widespread and is…

  • Vertex-AI

    Vertex-AI

    At the Google I/O developer conference earlier in May, Google Cloud announced the global launch of Vertex AI, a machine…

  • Axios Vs fetch what should you use

    Axios Vs fetch what should you use

    One of the basic tasks of any web application is to speak with servers through the Hypertext Transfer Protocol (HTTP)…

  • Async JS with Promises from Go

    Async JS with Promises from Go

    In JavaScript, Promise’s are the foundation of async/await. Lets take up an example, consider the below code, This will…

  • JavaScript Objects in Go and WebAssembly

    JavaScript Objects in Go and WebAssembly

    WebAssembly (abbreviated Wasm) is a binary instruction format for a stack-based virtual machine. Implement your…

  • JavaScript Design Patterns

    JavaScript Design Patterns

    A pattern describes a Challange that happens over and another time in the environment, so describes the core of the…

  • Top 4 Reasons to adopt JAMstack

    Top 4 Reasons to adopt JAMstack

    JAM stands for JavaScript, APIs, and markup. Primarily you build the principles for the generator in JavaScript, use…

社区洞察

其他会员也浏览了