Big Bang vs. Incremental Migration: Which Cloud Strategy Wins?
According to a report from Pluralsight, 94% of companies worldwide are adopting cloud computing, and experts anticipate cloud spend to reach nearly $1.2 trillion by 2027. With the rapid increase in the popularity of cloud computing, it is tempting for organizations to join the bandwagon and go all in. However, this “big bang” approach to cloud migration could backfire, leaving business operations and revenue vulnerable.
Enterprises that elect to shift to cloud migration in one sweep — the “big bang” approach — may risk severe consequences for their business, technology, and margins.
On the business front, a disorganized cloud migration could lead to service disruptions, upsetting customers and negatively impacting revenue. With regard to technology, rushing your cloud migration may leave little time for thoughtful considerations, such as appropriately matching service needs or future scalability. If businesses do not take the time to understand how the code operates, they may reinvent the wheel only to discover a solution that could have been implemented from the start.
Finally, a common belief is that moving to the cloud will reduce costs. However, cloud migration costs could quickly balloon beyond the intended budget if a company lacks a framework and practices to maximize business values, commonly known as FinOps. I learned firsthand the importance of implementing FinOps governance to operate smoothly at scale. With over 100 independently running teams at New Relic, we created a FinOps framework and implemented a weekly inspection mechanism. As a result, New Relic saved $20M in the last fiscal year.
This “lift and shift” strategy, which involves moving everything at once, is often driven by a mandate to have everything on the cloud. The magnitude of data to migrate can exacerbate the above risks and requires significant time, often six months or longer. However, another strategy that reduces risks while achieving successful migration is moving to the cloud piece by piece.
Why Companies Are Ditching the Cloud: The Rise of Cloud Repatriation
Testing the Waters With Incremental Adoption
Cloud migration is an ongoing, open-ended experiment for companies. Early in our cloud journey, I recognized that our migration would be a multiyear process. I led the team to reconfigure our platform for hybrid operations, a decision that positioned us well for multicloud environments.
So, how should enterprises get started? Starting with smaller projects allows companies to test different approaches, learn from the outcomes, and identify effective strategies and pitfalls early on. This iterative process helps them gain insights into cloud adoption, informing their overall strategy.
For example, the teams overseeing cloud migration can adapt and evolve with the process’s needs while carefully recruiting new talent to complement existing employees. Similarly, the FinOps team, which bridges the gap between finance and engineering, can continue to refine its practice, and legal and compliance teams would have the opportunity to ensure they can meet all regulatory standards.
Further, cloud migration presents a valuable opportunity for organizations to enhance and streamline their systems without becoming overwhelmed. For instance, as organizations move to the cloud, they can adopt auto-scaling to optimize costs while still meeting demand, implement enhanced network and security isolations between workloads to minimize risk and utilize availability zones (AZs) to increase overall availability. By establishing these strategies, companies can maximize the benefits of cloud migration while maintaining a more efficient and secure system.
Measuring Success at Every Step
With an incremental approach to cloud migration, companies can continuously track the process’s success and measure it against relevant KPIs. The first step is to benchmark the original baseline. What are the operational costs? How reliable and functional is it? What is the current satisfaction of key stakeholders, such as end users? Additionally, the company can set KPIs based on its desired cost, reliability, and satisfaction level.
After establishing the baseline, the team overseeing migration to the cloud can check the current status against those benchmarks throughout the process. While progress may not always be linear, it should at least trend positively. Likewise, the team can investigate and remediate the issue if the check-ins show concerning anomalies, such as a dramatic increase in cost or downtime.
Reliability: The Key to a Successful Migration
Cloud migration is often a multiyear process, and companies must be prepared to operate within that extended timeframe. While systematically moving workloads to the cloud, it is necessary to simultaneously maintain a connection to on-premises systems to ensure uninterrupted service for users. To effectively manage this complexity, it is essential to have effective visibility, allowing teams to identify and troubleshoot issues.
Whether a company employs a lift-and-shift or an incremental method, one question remains: How can you determine if something goes wrong? No process is infallible, and it is vital to amend identified issues quickly. The key lies in prioritizing reliability and leveraging a robust observability practice to effectively monitor problems and their underlying causes.
As the technology landscape continues to evolve, companies that adopt careful and experimental migration strategies with efficient monitoring are more likely to succeed without jeopardizing their operations or financial health.
