The key to successful cloud planning

Taking a step back from the all-in cloud approach

Patrick O’Keeffe
© Shutterstock / Allies Interactive

The cloud is a journey, not merely a destination. Keep this in mind as your organization moves to the cloud. Take a step back and re-examine your approach. Should some of your databases stay on-premises, and if so, which ones? When should you shift to a multi or hybrid cloud approach? This article discusses four potential approaches to take that are not just an “all-in” mentality.

As we look back, most agree that 2019 was the year of the cloud. Many companies embarked on an all-in cloud approach with 83% of enterprise workloads expected to be in the cloud by the end of 2020. Yet cloud migration is a journey, not a destination and this year we’ll see enterprises take a step back, re-evaluate, and be more strategic in moving to the cloud. While there are benefits, there’s also the effect of high costs and a lack of flexibility. Because of this, we’ll see more companies reconsider and change their approach to the cloud and move from an “all-in” mentality to a “hybrid or multi-cloud” approach, particularly when it comes to managing databases.

SEE ALSO: Serverless chaos engineering – interview with Emrah Samdan

The shift to multi or hybrid cloud

Moving every single one of your 1,000+ databases to the cloud sounds like a great idea. But as we all know, unexpected challenges and additional costs can quickly surface at any step of the journey – making the process of moving to the cloud much easier said than done.

There’s a typical pattern for companies embarking on a cloud strategy and it looks like this — companies will lift and shift infrastructure to the cloud, see costs increase, then move some workloads back to an on-premise environment. Then application lifecycle drivers push new apps (or new features for old apps) to be built in the cloud using PaaS/DBaaS technologies with more favorable cost models, then retire old IaaS apps. The key takeaway is that this scenario creates a need for a hybrid cloud approach.

Taking a hybrid cloud approach enables organizations to leverage the cloud (using AWS, Azure, or GCP) for some applications and computing needs while still keeping mission-critical or sensitive data closer to home. On the other hand, with a multi-cloud strategy, organizations can reduce costs and, instead of being constrained to one cloud, departments have the flexibility to select the service that works best for their individual needs (using a mix of AWS, Azure, or GCP). Multi-cloud enables organizations to spread the load for application workloads and performance, utilizing specific clouds for specific needs.

Back at square one

So how can your organization get started in determining which databases are better suited to be migrated to the cloud vs. remaining on-premises? When should you consider a complete transition to the cloud vs. starting small and transition only a portion of your database infrastructure to the cloud? While there are a few different approaches that have different pros and cons, make sure to evaluate which approach is best suited for your organization. Whichever path to migrating to the cloud makes the most sense for your organization, remember: planning is key.

What works for one organization may not work for another – and what works for one department may not work for another. With that in mind, here are four different approaches to database cloud migration that can help you get started:


Develop net-new applications using databases in the cloud, without migrating old databases at all (clean, but hard on historical data). Some companies are best suited to start from scratch, with no historical data on hand. This approach involves taking an on-premise database, setting it up in the cloud with every customization and configuration needed – then starting anew at the beginning of a new fiscal year or quarter.

The on-premise version can still persist in case there’s a need for queries on historical data, but not migrating the old database can make for an easy, light-lift to transition to the cloud. However, while it offers a clean switch, this approach means you will retain some of the hardware and software costs associated with the original on-premise databases.


Switch an entire on-premise database to become a database in the cloud (comprehensive, but requires protracted downtime). This is the “big-bang” approach that backs up and restores everything before users are affected. It can result in a quick migration, such as over the course of a weekend, or even overnight in some cases.

This approach involves interruption and risk, but moving the system while it’s not being used can work for applications with small databases and regular downtime, such as brick and mortar companies with regular business hours.

In this approach, DBAs back up the database and applications, restore them into the cloud, and start users on the new system effective the next business day. But, if the database measures in terabytes, it can take much, much longer than just a weekend to restore to the cloud. Overall, this can be an effective method to database migration – but it is considered a higher-risk and increases the possibility of extended downtime.

SEE ALSO: What it actually costs – an ecological price tag for software development


Identify low-impact tables and schemas to begin migrating to the cloud first (low risk, but slow). This is a good way to start small, and low hanging fruits to consider migrating first include development or QA databases. Other use cases, like data integration, disaster recovery and offloaded reporting that require data availability, are also good places to consider migration since these use cases don’t interfere with application uptime.

Moving with a piecemeal approach can be helpful to ensure applications and services aren’t disrupted – and it can show the benefits of the cloud to a broader audience. However, while this is a very low-risk approach, keep in mind that it is a slow approach that can take a long time to realize fully cloud-based applications and services.


Replicate data from an on-premise source database to a target database in the cloud (innovative). This approach can result in a safer, less-drastic way to ease into migration to the cloud from legacy and on-premise databases. It still allows for access to faster and cheaper compute, easier database management, and can help organizations reap the benefits of the cloud without necessarily having to dive all-in. With this approach, while tooling might be needed, it can provide a low-impact way of modernizing infrastructure quickly and affordably.

Whichever path to migrating to the cloud makes most sense for your organization, remember: the cloud is a journey and not a destination. And as we’ll see in 2020, with many organizations moving away from an all-in cloud approach, remembering this is vital. There are different benefits to be had from going the hybrid or multi-cloud path. Make sure to evaluate the needs of your organization and make decisions based on what will work best and benefit the most.


Patrick O’Keeffe

Patrick O’Keeffe is VP of Software Engineering at Quest Software, where he leads the software engineering team for the Quest Information Management business unit. Patrick has over 25 years’ experience in leading software engineering teams, software engineering and architecture and is based in Austin, TX.

Inline Feedbacks
View all comments