Originally Published in Forbes.
Shailesh Manjrekar is the Chief Marketing Officer at CloudFabrix, the inventor of Robotic Data Automation Fabric and an AIOps Leader.
For the last couple of years, I have been highlighting yearly predictions at the beginning of each year. For 2023, I would like to expand on one of the trends that I briefly mentioned in my 2022 predictions: superclouds.
Since then, there have been a lot of discussions, vendors staking their claims, and clarity around superclouds. In this article, I will look at how observability, AIOps and automation are going to be key tenets for supercloud abstraction and adoption in 2023.
What Is A Supercloud?
As I explained last year, superclouds are domain-specific clouds built on top of hyperscalers to derive deeper business insights, better time to market and innovation. In simple terms, a supercloud is a layer of abstraction across edge clouds, private, public and multiclouds.
A supercloud can help address the challenges associated with multicloud in terms of application portability, abstracting APIs, data formats and vendor-specific IaaS and PaaS details. Not only are cloud and software vendors (Snowflake, Databricks) beginning to build their own superclouds, but companies outside of the tech space (Goldman Sachs, Nasdaq, United Airlines) are creating their own as well.
Businesses have been adopting multicloud for a number of reasons, such as preventing vendor lock-in, controlling costs, improving inheritance due to acquisitions and improving expertise in certain verticals like Google Cloud Platform with OpenAI. It has also been tasked with mitigating the effects of the rise of edge clouds and IT and OT convergence with Industry 4.0. However, multicloud has brought with it tools sprawl as well as increasing difficulties with observability and AIOps due to the nonstandardization of APIs, data formats, infrastructure and cloud services.
Superclouds have the potential to overcome the challenges of multicloud by standardizing access, portability and governance and eliminating data silos. With the current macroeconomic climate, businesses will likely focus on their core domains and digital transformation is going to be about driving an experience economy. One of the ways to achieve these goals is with the supercloud, which has the potential to improve simplification, time to market and customer experience and cost economics.
The Journey To Supercloud
Supercloud adopters will fall largely into two camps.
1. Customers innovating new business models that are edge-centric and cloud-native.
2. Customers who have started their hybrid and multicloud journeys but are struggling to simplify their deployments.
The “cloud operating model” is now well understood, and it enables a customer to move from a static, siloed, on-demand data center to an on-demand, ephemeral, integrated data center where infrastructure is deployed as code, networking is service-driven, security is identity-based and applications are containerized and deployed as CI/CD pipelines.
The AIOps operating model, on the other hand, may be more useful for the journey to supercloud. Companies will need to unify observability for visibility, and AIOps can help to infuse AI/ML pipelines with MLOps and automation for DataOps to resolve data quality, skills gaps, data silos and ServiceOps. The AIOps operating model is anchored on platform engineering teams and self-service personas to provide governance, simplicity and agility for superclouds.
In a recent post, VMware explains five categories of supercloud, or what they are calling cross-cloud services: application, infrastructure, security, end-user and data plane services. The AIOps operating model can help enable observability, AI/ML and data and ServiceOps automation across all these five categories.
The AIOps operating model ties the operational and developmental functions to a profit center, with the direct impact on focusing on enabling and accelerating the business. The AIOps operating model thus needs to span across people, processes and resources in order to implement these changes.
• People: Self-service-based agile principles can be enabled—and the skills can potentially be mitigated—by delivering low-code/no-code environments.
• Intelligent Services: The AIOps operating model requires composable services infused with data and AI/ML.
• Resources: Finally, it requires optimizing resources (systems and software) with cost optimization, operational excellence, performance efficiency, reliability and security.
By focusing on these goals, the AIOps operating model serves three high-level objectives.
1. Democratizing data-first, AI-first and “automate everywhere” strategies.
2. Providing a blueprint for embarking on an automation journey by operationalizing edge, hybrid and multi-cloud deployments.
3. Quantifying the value and economic benefits of using AIOps.
In short, we should see the birth of many more superclouds this year. Those looking to take part in this trend and operationalize superclouds should explore how to adopt observability, AIOps and automation in order to streamline their supercloud journey.