Modernizing your Integration infrastructure and applications
This blog is an attempt to distil the Cloud Adoption Frameworks (CAF) from the three largest cloud and service providers (CSPs) i.e. AWS, Azure Cloud and Google Cloud, and provide a combined framework based on the understanding of what each phase contributes to the process and their inter-relationship. I am also taking this opportunity to investigate how the Cloud Center of Excellence (COE) complements the CAF and determine if the CAF application to an integration workload would be any different from a traditional workload.
CLOUD ADOPTION FRAMEWORK
For organisations who have embarked on the journey to the cloud, the cloud adoption framework or playbook would be something that they are familiar with. The framework provides detailed guidance for a business organisation looking to migrating their workloads to the cloud. A typical framework consists of the following phases (executed in the order listed).
The Cloud Adoption Framework focuses on helping you understand your organisation’s current cloud maturity state and the transition required to achieve the target state. To help your organisation reach the desired cloud maturity level, the CAF highlights a set of clearly defined, nonoverlapping workstreams aligned to your stakeholders.
The framework affects the business, technology, and culture of enterprises. Sections of the Cloud Adoption Framework interact heavily with IT operations, IT governance, finance, line-of-business leaders, networking, identity, and cloud adoption teams.
Below is a brief description of each of the phases. The framework is not prescriptive. You can pick and choose the stages and activities to adopt depending on your needs, the workload, and the amount of investment you are prepared to put into the framework. You may decide to start small and slowly increase functionalities and governance as more workloads are migrated.
STRATEGY
The aim of the strategy phase is for the business to document and communicate its business strategy. Capture the motivations behind cloud adoption and the expected business outcomes. This information allows you to match the cloud adoption strategy to specific cloud capabilities and business strategies.
PLAN
The planning phase happens when the adoption strategy is realised as a concrete plan of action. The organisation’s digital estate is identified and documented and, organisational and skills gaps are addressed with plans. Cloud adoption plans are drawn to manage the digital estate, skills, and organisation changes.
READY
The ready phase gets you familiar with the target cloud setup and tools to create a cloud landing zone for your workload environment. The landing zone configuration is validated against best practice as your cloud adoption evolves.
MIGRATE
Leveraging the CSP’s migration tools and best practices, start migrating your workload using one of the following approaches.
- Lift and Shift
- Lift and Optimise
- Modernise
- Relocate
- Repurchase
Regardless of the approach, each workload goes through the three phases of migration:
- Assess workload
- Deploy workload and
- Release workload (to Production)
INNOVATE
Besides focusing on the migration and modernisation of existing workloads, cloud adoption-related innovation can unlock new technical skills and expanded business capabilities. The deployment of containers and adoption of serverless technology are some of the ways to further optimise your business values.
GOVERNANCE
Build new cloud governance processes and policies to complement the existing IT on-premises governance as the digital estate in the cloud grows and evolves. Iteratively add governance controls to address substantial risks as you progress toward the cloud adoption end state.
MANAGE
Plan for reliable, well-managed operations of the cloud solution is critical to the ongoing operation of the digital assets that deliver tangible, valuable business outcomes.
REVIEW YOUR OPTIONS
To fully understand the target audience, activities, and deliverables in each phase, I strongly recommend that you spend some time reviewing one of the public cloud vendors’ cloud adoption framework.
The above are the phases outlined in the Microsoft Azure Cloud Adoption Framework (CAF). The Google and AWS CAFs use different terminologies and group the stages differently, but they distill loosely to the same three areas with similar approaches.
- Business
- Technology
- Process
All the major CSPs also recommend a suite of tools to facilitate cloud adoption. Some of the useful aids from the various vendors are:
- The AWS Action Plans
- TCO calculator
- Google Cloud Maturity Scale
- Azure Blueprints
- Azure Strategic Migration Assessment and Readiness Tool (SMART) workbooks and templates
These tools can help an organisation migrate to the cloud faster, identify gaps, realise their desired outcomes, and avoid common challenges to a smooth migration.
Once the cloud maturity goals are defined, create success metrics and align them with your strategic business goals.
CLOUD CENTER OF EXCELLENCE
An organisation should also consider complementing the CAF with a Cloud Center of Excellence (COE). The goal of the Cloud COE team is to understands the CAF and to use it as a guide for implementing cloud technology aligned with a business’s goals and strategy.
The Cloud COE team effectively becomes the channel for transforming the way that the other teams serve the business in the shift to the cloud. The Cloud COE team accelerates cloud adoption by:
- Driving momentum across the organisation
- Delivering quick wins
- Developing reusable frameworks for cloud governance
- Managing cloud knowledge and learning
- Engaging and evangelise
- Overseeing cloud usage and plans for scale
- Aligning cloud offerings to the broader organisational strategy
Tracking the key performance indicators (KPIs) allows the management and Cloud COE team to align communications, upskilling, and practices to help employees incorporate the cloud into their work.
MIGRATING INTEGRATION SERVICES TO THE CLOUD
With a good understanding of what is involved when migrating to the cloud, let us consider the steps needed to migrate your integration services to the cloud. The data integration services are usually part of larger application workload, e.g. Big Data or Machine learning workload. They may also be part of an application workload integrating the mobile services from various on-premises applications and cloud-based APIs.
Once you have strategised and developed a plan to move your integration services to the cloud. You are now ready to begin the prep work before the actual deployment or migration of your workload to the cloud can occur. In this Ready phase, you are building out your cloud landing zone (aka environment) for the selected workload in your CSP-hosted environment. This phase is also the time when you start to consider the system requirements, recorded as part of the digital estate in prior stages, and what is available on the CSP offering to build your landing zone to reach your initial target maturity state. You may consider a strategy of starting with a Minimal Viable Product (MVP) type landing zone and building that out as you gain more experience and the requirements increase.
Areas that you need to consider and prepare include:
- Organise resources
- Manage access
- Manage costs and billing
- Plan for governance, security, and compliance
- Use monitoring and reporting
You may also decide to adopt a platform to help address some of the environmental workload requirements, minimising any remediation and enhancement work further down the track. An example would be to land on a managed container platform like Red Hat OpenShift or one of the public CSP’s native managed container or serverless platform. Some CSPs offer deployment blueprints that help you build out the environment based on their best practices simplifying the preparation tasks and ensuring it is following best practices.
We can begin the adoption next.
For the selected workload, the effort to deploy may involve minimal changes, consider the following migration strategies:
- Lift and Shift
- Lift and Optimise
- Modernise
- Relocate
- Repurchase
Adhere to the public CSPs’ migration guide and best practices checklist to ensure that you are leveraging the right tools to achieve the required outcome while conforming to the vendors’ recommended approach. Some even created common migration scenarios with detailed instructions.
For a COTS software product, like an integration platform, migration to the cloud may be as simple as the following the migration approaches mentioned above assuming the software platform was set up and configured as part of the readiness phase – i.e. it forms part of the landing zone. We are planning to provide a custom migration scenario to cover IBM Cloud Pak for Integration platform in a future blog.
Once the integration workload is running in the cloud, use the cloud-native tools available to manage its running cost, sizing and refactoring the solution. Learn how to leverage the advisory services offered by the cloud vendor to help you secure, optimise the workload performance and cost. The optimised workload is ready for promotion to production. It is always a good idea to engage the cloud vendor services team or a partner to assist and validate the approach at various stages of the migration exercise.
With the workload deployed, management of the cloud operation is vital to ensure the business outcomes registered in the planning phase can be realised. A management baseline is highly desirable with:
- Clear visibility of the running state of all deployed assets
- Proper process and procedures in place for each workload
- Baseline security and backup controls in place
To reach the desired cloud management maturity, an organisation needs to onboard new tools and services to manage the cloud operations. Depending on your cloud footprint, it may involve multi-cloud management and existing on-prem tools to cover hybrid deployments. The cloud management baseline is also extendable to cater for mission-critical workloads that require specialised application insight and operational support that may reside outside of the cloud operation team responsibility.
Existing governance framework will also need to be reviewed to incorporate the new digital assets. New policies have to be implemented while ensuring they complement existing policies. As your cloud maturity and footprint grows, your cloud governance will undoubtedly evolve. For an organisation starting its journey in the cloud consider an initial governance foundation that aligns with the business goals and strategy. Establish the boundaries and guardrails for your digital assets to reach the target business goals using the documented business tolerance, risks, regulatory requirements and the characteristics of the applications in the strategic and planning phases.
The formulated governance policies normally land into the following disciplines
- Cost management – ensures cloud expenses are within budget, the workload is sized correctly to avoid wastage, abnormal cost spikes are managed, and reserve capacity is not wasted
- Security Baseline – attempts to address core security-related business risks starting with data security, service availability and access control
- Resource Consistency – ensure that resources are deployed, updated, and configured consistently in a repeatable manner, and that service disruptions are minimised and remedied in as little time as possible
- Deployment Acceleration – ensures that your cloud resources are deployed, updated, and configured correctly and consistently, and remain in that manner. Coupled with deployment automation, it indirectly affects your cost management strategy
To assist an organisation ramp up their cloud operation management and governance, all the CSPs have developed a suite of capabilities and tools that can be leveraged to provide those functions. We have also seen some of these functions in container platform offerings that complement the CSPs’ offerings an example is Red Hat OpenShift Container Platform.
SUMMARY
In summary, the considerations and process of migrating integration services to the cloud are not dissimilar to any traditional application workload. The journey to the cloud may be different for each organisation, but there are signposts and beaten paths that help you reach the outcome you desire based on your business objectives. All major CSPs are keen to assist with your cloud adoption and based on our review of the CAFs available, Microsoft Azure is the most comprehensive and complete with templates, blueprints and tools to help you on this journey. AWS has a prescription guide, based on its Well-Architected Framework and its complementary migration tools, is a well-trodden path to the cloud. To create widespread advocacy and accelerate cloud adoption, the Google Cloud’s Building a Cloud Center of Excellence Guide is a good place to start.
REFERENCES
- Microsoft Cloud Adoption Framework for Azure
- AWS Cloud Adoption Framework
- Google Cloud Adoption Framework
- Google Cloud – Building a Cloud Center of Excellence
- AWS Cloud Center of Excellence Blog
- Azure Cloud Center of Excellence
- AWS Prescriptive Guide – Migration Strategy