As companies look to move workloads to the Microsoft Azure public cloud platform, having a well-constructed strategy is critical to realizing the transformative power the cloud offers.
Making the first move to the cloud can be intimidating. If an organization starts without the right planning, or without following best practices, the results can lead to a negative cloud migration experience and business setbacks.
These steps are important to planning a successful Azure migration that results in a robust cloud deployment that’s closely aligned to your business goals.
Preparation and Planning
Every cloud migration begins with planning. You’ll need to get the right stakeholders involved, determine the ideal timeframe for completing key steps, allocate budget, and then get ready for the move.
Defining Scope and Success
Identify workloads and projects that may be good candidates for moving to Azure, then define the scope of the migration including costs vs budget and priorities. A poorly defined scope, workload or project can lead to problematic and stressful migration. An experienced partner – CSP or consultant – can offer guidance on identifying the right projects and keeping the scope tightly focused.
- Start by comparing applications or products that are supported by legacy technologies (older operating system versions, older source code versions) vs applications or products supported by modern technologies.
- Keep the scope in focus by performing thorough discovery at the front end of the project to fully understand and identify all necessary resources and components to support a product or application that will be migrated. Doing so will enable project managers to understand whether project changes that may be requested are necessary for project completion.
Define milestones and project success during the migration planning phase to have a more positive experience. When there are goalposts in sight, teams maintain focus and are better at avoiding scope creep.
Best Practice Tip: Understand your team’s work pattern. If there are services that can be shut down during off-hours, it is highly recommended to build a flexible architecture that takes this into account. Azure charges on a utility model so cost savings can be realized by only running VMs when they are needed. Leaving instances running when they’re not in use can lead to unexpected costs on your next cloud bill.
Creating and Sharing Standard Operating Procedures (SOPs)
Standard operating procedures should be fully built out and understood by all parties prior to the migration project. When partnering with a CSP, existing SOPs and runbooks can be reviewed and updated, based on the experience the partner brings to the project.
Best Practice Tip: Take advantage of cross-team training on SOPs. This produces better SOPs in the end as it requires them to be highly understandable and procedurally accurate.
Identifying and Planning for Complexities
Many enterprises have to consider data sovereignty and compliance requirements. Understanding the implications of moving to Azure in light of those factors can be daunting.
Eliminate surprises from coming up during the migration phase by identifying the workloads that need to be updated or reconfigured in order to function properly in Azure in the preparation stages.
Best Practice Tip: Bring InfoSec teams into project planning at an early stage. They often have a perspective on design and migration processes that other teams may not have considered.
Once your preparation is done and goals are set, you’re ready to determine your migration strategy.
Architecture Solution Designing
There are many factors to consider when designing end-state architecture. Azure offers multiple service options that should be understood to make the best-informed decisions.
- IaaS vs PaaS – how much control do you want to have over your cloud infrastructure?
- Choice of database platforms – SQL, MySQL, PostgreSQL, and Cosmos DB are all natively supported and have PaaS options available.
- HA and DR design options – building for business continuity. Decisions regarding high availability and disaster recovery often hinge on a budget, but Azure offers multiple paths to higher resiliency. Taking advantage of Availability Sets, Availability Zones, Azure Backup, and Azure Site Recovery are all cost-effective options.
Best Practice Tip: Traditionally, infrastructure planning focuses on long-term growth so systems are provisioned to handle expected growth over multiple years. Azure presents a new paradigm: provision systems to handle current load at near capacity and expand when demand dictates to get the most bang for your buck.
Choose a migration path… depending on the architecture solution necessary to house and run the workloads, along with the project scope and complexities. Your choice is between shifting applications directly into the cloud or rebuilding to maximize functionality.
- Lift and shift – if resources are cloud-ready, they can be ported over to a cloud environment. This can save on upfront costs but may result in lower efficiency down the road, because the apps aren’t designed to take full advantage of cloud capabilities. Also, Azure updates and optimizes images frequently, so using Azure images keeps you running on the best platform available.
- Build and migrate new data – when there are more complex environments or legacy workloads that need to be upgraded or reconfigured for cloud readiness
Best Practice Tip: If possible, try in-place upgrades of applications running on older operating systems. This will help determine cloud-readiness and help make decisions around the migration strategy as you gain insight into stability and performance after upgrades.
Application-specific migration paths
- Lift and shift – moving existing infrastructure directly into the cloud. Best suited for applications already hosted on modern operating systems and infrastructure
- Re-platform – deploying existing code onto modern infrastructure with minimal changes. Best suited for applications that are currently configured with legacy technologies but are known to be compatible with modern technologies
- Refactor – reconfiguring applications for cloud-readiness. Best suited for applications that run on legacy technologies and cannot be run on modern technology as currently configured.
Best Practice Tip: Azure App Services offers highly resilient, portable and cost-effective platforms for application hosting. Whenever possible, look to migrate out of VMs and into App Services.
What Happens Next?
A well-architected solution will enable an organization to migrate to the cloud, and then grow and scale for years. Choosing the right type of infrastructure to build and choosing a smart migration path are essential. But the Azure environment also has to be maintained and managed to continue to operate at the highest level.
Companies that successfully use Azure also keep up with active monitoring, security, alerts, cost control, rightsizing and ongoing planning to keep their cloud aligned with changing business goals.
Learn more about how Atmosera can handle your transition to the cloud.