Where Innovation Converges with Efficiency: The Cloud Continuum

Cloud technology has been transformative, driven by the cost benefits and flexibility of the public cloud, resulting in rapid and widespread adoption. As public cloud usage grows, enterprises must look at the whole landscape as a "cloud computing continuum" spanning across centralized public and private clouds, as well as decentralized edge and far-edge environments, to balance innovation and efficiency.

Analysts predict a significant shift towards decentralized cloud environments for enterprise data management. According to a report by McKinsey, by 2025, a vast majority of enterprise data will be generated, processed, and analyzed in decentralized environments. McKinsey also states that AI demand has spurred significant growth in cloud and edge computing, with the Cloud AI market anticipating a 30.9% CAGR from 2023 to 2030. Companies are boosting cloud spend by 20% to enhance model training and inference. Efficient workload distribution across cloud and edge optimizes resources, latency, data privacy, and security, unlocking business value.

Challenges in Adopting Cloud Continuum

Enterprises today must adopt a nuanced cloud strategy that effectively balances performance, reliability, innovation, and cost efficiency. One destination is not viable, instead a stable and well-managed cloud computing continuum is essential for success.

The enterprises should assess whether a workload genuinely benefits from public cloud features such as elasticity and scalability. It’s important to consider data gravity – data dependencies between applications – and the investment viability when deploying applications in the public cloud. For instance, a logistics firm encountered performance issues during cloud migration due to incorrectly assessing data gravity among its applications, which required a temporary pause to refactor the interfaces between applications before resuming their cloud journey.

Also, skills and tools required to manage various deployment zones in the continuum could vary which could affect the overall business service assurance.

Analyzing Workload Placement Across the Continuum

Workloads are categorized into three horizons, each representing different levels of stability, agility, and innovation, that determine how they are placed and managed across the continuum.

Horizon 1 consists of existing core business applications that are stable and meet both current and future needs, typically comprising 50-60% of workloads.

Horizon 2 includes rearchitected or new capabilities developed with agile principles to support business growth and agility, making up 25-40% of workloads.

Horizon 3 encompasses new digital capabilities developed like start-ups to drive business disruption, accounting for 10-15% of workloads.

Private clouds are suited for Horizon 1 applications as they are built on standardized technologies and have relatively static workload characteristics. They also address data sovereignty needs more effectively. In contrast, public clouds provide fast access to new services and are scalable, making them more suited for the rapid development and scaling of Horizon 2 and 3 applications, including AI workloads. A public cloud-first strategy might involve migrating Horizon 1 applications to the public cloud, as new Horizon 2 and 3 applications often depend on the data and processes of the former.

Decision-Making Framework for Cloud Continuum

To make informed decisions, it is essential to evaluate the "4Cs" – capability, compatibility, cost, and compliance – to determine the right deployment zone for the workloads. This necessitates a seamless experience for developers and operators across the continuum, which can be delivered by adopting cloud platform engineering. By creating a “resilient platform” with a unified control and management plane, and leveraging automation and AI tools, software engineering, delivery and operations can be effectively optimized.

The Future of Cloud in Digital Transformation

Generative AI and Emerging Technologies: Generative AI is reshaping cloud strategies by providing new avenues for innovation. While the public cloud is often the preferred environment for AI Proof of Concepts (PoCs) due to GPU availability, selecting the right destination in the continuum is crucial as these projects move to production.

Regulatory Considerations: Regulatory requirements, such as the European Banking Authority's exit strategy guidelines, complicate cloud strategies. Building the target state as a cloud computing continuum and building capabilities to transition workloads between public and private clouds can help meet regulatory needs. Additionally, incorporating automation and AI in compliance checks and control implementations can make managing the continuum easier, more cost-effective, and compliant.

SaaS Evolution: As more commercial off-the-shelf (COTS) applications transition to SaaS and offer generative AI for a co-pilot experience to business users, enterprises innovate faster through highly productive business applications with the SaaS version of the product. This reshapes cloud strategies for the enterprise as SaaS becomes a crucial component of the cloud computing continuum.

The Future – Being Always Right in the Cloud Continuum

Regularly reassessing workload placement is crucial to keeping enterprise cloud strategies aligned with evolving business needs, technological advancements, and regulatory requirements. As every major enterprise drives innovation with AI and generative AI technologies, there is also a need to focus on efficiencies to ensure profitable growth.

Written By - Madhan Raj J, Associate Vice President, Cobalt Cloud Solution Strategist, Infosys

Read More:

Cloud Server Security Risks Affecting B2B Domain