Product-based, service-oriented architecture is more relevant than ever. Businesses are increasingly realizing the need to have a modern, agile approach to software development and infrastructure architecture. The key challenge, however, is building a platform solution that can deliver on the power of service-oriented architecture, while providing a foundation to meet future cloud demands. 听A few questions that need deliberation are:

• How is our industry changing and what short-term and long-term needs will need addressed?
• How can the company efficiently address current cloud needs while establishing a framework for future growth?
• Can our platform solution be used with application servers, development tools, etc. from different vendors or will it force vendor lock-in?
• Does my current infrastructure support agile growth with enough flexibility to quickly pivot to changing business needs?

Based on the key challenge and the self-assessment questions, businesses can arrive at a new way of thinking. A Cloud Centre of Excellence and Expertise (CCoE) leads the organization in cloud adoption, migration, and operations. It is a cross-functional ecosystem of people, processes, and tools embedded in a business-aligned operating model. CCoE is responsible for developing and managing the cloud strategy, governance, development, and best practices - which the rest of the enterprise can transform the business leveraging the cloud.听

The Cloud Centre of Excellence (CCoE) consists of three pillars:

• Cloud Business Office (CBO) focuses on people and drives architecture, standards, business objectives, relationships, and guidelines.
• Cloud Central Services (CCS) centers on technology engineering, professional services, and operations improvement.
• Cloud Migration Factory (CMF) encompasses the industrialized capability that delivers cloud transformation as a service (TaaS).

CCS challenges

Of the three, Cloud Central Services (CCS) offers the most challenges since it deals with multiple personas trying to understand, use, and modify the infrastructure-as-code (IaC). Also, with the constant evolution of governance, compliance, and security, it can be daunting to balance best practices, while still promoting lean, efficient delivery.

Businesses may also have skilled operators who act as producers, managing cloud infrastructures, and developers who abstract themselves from provisioning infrastructure to avoid being the impediment to delivery.

The disparity of tooling and processes increases time and resource costs, while different teams us different tools and methods, making team coordination difficult. Mixing ad-hoc development workflows and disparate infrastructure deployment processes exemplify disorganisation.

Evolving to a new operating model

A new target-operating model is evolving for the cloud to satisfy product-based service-oriented architecture. The model combines a platform engineering team, enriched with site reliability engineering capabilities that is responsible for operations. This team uses templates, IaC scripts and policies, to deliver capabilities to product teams, who consume services via APIs.

A horizontal platform can be implemented and deployed immediately , enabling developers to build reusable interfaces while establishing a core framework for integration. Enterprises no longer need to rip and replace .

Platform and Cloud Infrastructure Services - New Target Operating Model听

• Responsible for continuous product design, build and implementation
• Product-level defect fixes听
• Architecture and engineering maturity improvements
• Objective: Agile and productive, high value

Platform Engineering Squads:

• Responsible for building and maintaining discrete IT infra products & services (API鈥檚, Service Catalogues, O/S, PaaS, Containers)
• Readymade Infrastructure as a Code (IaC) and secured policy as a Code (PaC) scripting
• Responsible for end-to-end Service monitoring, highly multi-skilled, leveraging Rapid Ops, DevSecOps, SRE and Infra skills
• Discovering patterns from incidents & problem records to develop a service automation backlog
• Coordinating periodic upgrades of the underlying platform
• Service readiness checks and release management听

The nirvana stage is reducing and eliminating repeated manual tasks and fully automating the tasks as cloud adoption process matures.

Business outcomes:

• Productivity improvement听
• Reduction in deployment processes and costs
• Reduction of operational efforts and human interaction/errors听
• Enhanced demand management and standardisation of operations
• Better compliance, security guard rails and non-linearity

When the model matures, architects bridge technical and the non-technical stakeholders and implement standards that everyone can understand and apply.听The model brings transparency to everyone, supports growth, and provides a standard framework for common tooling across development, operations, and stakeholders.

Key outcomes:听

Improved visibility and collaboration

• Teams: Each team has different roles when it comes to working on cloud infrastructure. Integrate by specifying roles, creating specific teams, and controlling who has access to what.
• Projects: Once teams are created, projects can be assigned.
• Environments: Well-defined projects can be broken into manageable environments. Environments represent day-to-day work and should be specific and detailed enough for users. Descriptions can provide clarity.
• Architectures: the final layer is where teams can find cloud infrastructures they are working on with visibility on what the architecture is solving and the status for each.听

By having design, code, and documentation on the same platform, everyone can understand how the infrastructure is built. Teams can review code and suggest improvements. They can learn to reuse automations in other applications, troubleshoot, and review/audit for compliance.

After everyone has been on-boarded and cloud assets have been centralised, efforts can be undertaken to share a common vision on who needs to be responsible for what, along with a strong focus on execution.

Reduced costs and efforts spent on on-boarding and training new employees.

Once teams share the same working environment, communication and collaboration becomes easier. By simultaneously working on the same infrastructure, teams can mentor the new hires and teach best practices.

Teams can access infrastructure documentation and reuse templates created by other team members. By process end, the business will have a system to track all engineering work in one place and communicate it to stakeholders with minimal overhead.听

Going forward, build effective, remote distributed agile teams, with transparency and ways of working that are both flexible and scalable, and fuel product-oriented architecture that leverages a resilient and fully automated platform.

听