5 Things You Need to Plan for When Migrating Data to the Cloud or Back Out of It

In this special guest feature, Caio Milani, Director of Product Management at MarkLogic, discusses the need to have an exit strategy when embracing cloud technology in the event you decide to get the data out of one cloud and move it to another, or move it back to your own data center. He also offers 5 things you need to plan for when migrating data to the cloud or back out of it. Caio is responsible for various aspects of the product including infrastructure, operations, security, cloud and performance. Prior to joining MarkLogic, he held product management roles at EMC and Symantec where he was responsible for storage, high availability and management products. Caio holds a BSEE from the University of Sao Paulo and a full-time MBA Degree from the University of California at Berkeley.

No doubt you’ve heard about the potential benefits of moving applications and workloads to the cloud: cost savings, increased agility, greater scalability, etc. And indeed, your organization might already be leveraging cloud services for some of its business applications. But what if you decide to switch cloud service providers, or even bring some of your applications or other IT components back in-house for whatever reason? Planning a cloud strategy doesn’t just involve pushing data into the cloud and then moving on the next task. You also need to have an exit strategy, in the event you decide to get the data out of one cloud and move it to another, or move it back to your own data center.

Here are some considerations for making sure you have the ultimate flexibility when it comes to managing your data, and your cloud strategy.

1. Build a multi-cloud architecture

The saying “don’t put all your eggs in one basket” certainly applies here: don’t put all your stuff in one cloud. In order to be truly independent and avoid vendor lock-in, you might think about moving some business applications to one cloud provider, others to another, and relying on yet other providers for disaster recovery and application development. Maybe the Amazon Web Services (AWS) public cloud is best for your software-as-a-service (SaaS) applications, given that it’s the largest cloud provider and provides access to a large number of SaaS applications that work together. If you are already a Microsoft customer, run on Azure. Even if you are not, consider Azure for the value of the Azure stack. Consider swapping out various projects among cloud providers to avoid getting locked in.

 2. Prepare for end-of-service scenarios with your cloud providers

Relationships with cloud providers don’t always go well or end well. There have been cases where cloud providers made life difficult for clients in terms of getting their data back once the relationship came to an end. This can result in clients having to deploy costly solutions or time-consuming processes for recovering its data—or even in the loss of data if the right precautions are not taken when forging a service agreement. Make sure there is a clearly stated clause in the service contract relating to the safe return of your data, in the specific format, transfer mechanism, and timeframe that works for both organizations. Also, when designing your cloud hosted capabilities, expect that data and associated metadata, entities, and privileges are application independent.

3. Ensure independent access control of your data

It’s important to have access control of your organization’s data be independent of the cloud provider. For one thing, it helps to avoid issues related to getting back your data once the relationship ends. It also simplifies things when you switch cloud providers, or when the cloud provider switches subcontractors. Furthermore, external access control along with data encryption means that when you move data, the information remains encrypted and is safe even if it falls into the wrong hands. Maintain direct access to the back-end database and Web services that support data to enhance future deployment flexibility. Finally, creating your own custom applications and using a cloud provider mainly for infrastructure services can help your organization keep control of its data.

4. Leverage independent third-party tools and SLAs

Are you relying solely on data quality and data governance tools provided by the cloud vendor? If so, not a good idea. Make sure you use the best tool for a given task, whether it’s provided by the cloud vendor, a third-parry vendor or your internal development efforts. You want to consider avoiding the use of auditing tools from the cloud provider altogether, given that the provider might not have any incentive in auditing possible inefficiencies in its own services. Consider external audit logs that are under independent control, encrypted, and centrally accessible.

Service level agreements (SLAs) ideally should also be vendor neutral. Make sure you have independent tools to monitor your data as well as SLA metrics, and consider storing records of access logs and SLA metrics in a secure, external database.

5. Deploy cloud-neutral technologies

Keep your options open when it comes to where you can run software. Many software license agreements are specific as to whether applications can run in the cloud. Ideally you should have a fluid agreement that allows you to run the software wherever you want. Also, your applications should be able to run in any cloud or across a variety of cloud services at the same time, as well as on on-premise systems. The software should also support multiple data storage formats and industry standards, replicate data easily across environments, provide its own access control independent of a cloud vendor, and have a record of no downtime during data center migrations.

One of the benefits of the cloud is that it gives organizations greater flexibility, enabling companies to store and run applications and workloads from anywhere. But sometimes the cloud can actually create inflexibility.

By adhering to these five practices, you can help ensure that moving to and from a variety of cloud services remains a relatively unencumbered experience.

Sign up for the free insideBIGDATA newsletter.