Nine Tips for Choosing Cloud Data Migration Software

Key considerations for choosing a cloud data migration software for the near and long-term

As your organization strategically moves applications, architectures, and data platforms to the cloud, determining the best tool to perform your live data migration can be nuanced and complex. You may be tempted to simply choose a free option from your cloud vendor, who has a vested interest in seeing your data migration succeed sooner rather than later so they can start the meter. The free option built into your on-premises database might also look tempting. While it may not get you fully migrated, your talented team can probably manually engineer the other 20 percent.

But when planning your tactical execution, it pays to keep your long-term goals in mind.

Cloud data migrations are becoming increasingly complex.

Entire application and data stacks are not only being relocated but are being simultaneously rearchitected with clear, and sometimes not so clear, visions of the future. You likely already have one or more different applications or testing frameworks in one or multiple providers’ cloud environments (e.g., IaaS, PaaS, and SaaS). If not, you’re about to.

If you select the free tool for your current platform, you’ll likely end up with numerous migration tools that all work differently, do not integrate with one another, and have limited flexibility in your modern cloud environment.

According to a recent report by Gartner[1]: “Using multiple cloud service integration platforms… would rapidly create a chaotic environment made of multiple, redundant integration platforms. There would be significant overlap in functionality, and obvious cost, manageability, and skills implications. With each additional platform added to the mix, the challenges become increasingly difficult to address.”

Gartner concludes that “Ideally, you need a single platform that can run integration flows across on-premises and diverse clouds, thus providing a single integration development toolset and a single control pane.”

So, what should you look for in a cloud data migration solution that will address all your data migration needs moving forward?

Here are nine questions for you to consider:

  1. Does the migration software support a wide array of data endpoints?

Organizations rarely run all their applications on a single data platform. Most run several different applications on different databases and clustered file systems. If you want a single tool to support your needs, that tool should replicate between a wide range of database, data lake, data streaming and cloud platform types.

HVR supports a variety of cloud endpoints.

Supported_Platforms
  1. Does the software work with all the major cloud vendors?

Increasingly, organizations are using multiple cloud platforms. Some organizations are becoming reluctant to put all their applications and infrastructure eggs in one provider’s basket. Others may surgically place specific workloads where they make the most sense.

For example, they may place general computing, relational databases and file storage in AWS; business applications and SQL Server in Microsoft Azure; and AI (artificial intelligence) in Google cloud. The data migration solution should work consistently across all major cloud vendors for the sake of simplicity. You can earn one tool for all use cases – and avoid vendor lock-in.

 The data migration solution should work consistently across all major cloud vendors for the sake of simplicity

  1. Is the data migration software easy to set up and use?

Cloud adoption can be complex and daunting. The last thing you want is to spend months learning to use a data migration tool for a replication project that will take weeks to complete and then do it all over again for the next migration. So, the solution you select should be easy to set up and use.

Look for a solution that allows you to manage your entire set of data migrations with a single tool on a single console with a single look and feel. This tool should provide comprehensive migration capabilities to enable you to manage data table creation, initial load, synchronization, monitoring, data validation, testing and deployment. It should also minimize the number of clicks necessary for each operation.

  1. Does it provide the flexibility to do phased migrations?

Usually, when you migrate to the cloud, you’re moving an application and the database that supports it. Transferring the application can take considerable time because you must move the entire stack, including web servers, application servers, load balancers, and so on. Yet it’s the database transfer that causes the most downtime for users. This downtime occurs because data traditionally has needed to be offline when moved and all of it must be moved before it can be made available to users and updated again.

To minimize interruption, the solution should allow you to migrate the application stack first and then migrate the database. Using this phased technique, you need only keep your application offline during the database migration, which can take hours or days. The overall migration, which might take months, should be online throughout the migration process.

Another way to minimize downtime is to move your applications and users to the cloud gradually in a phased migration, rather than all at once.   Look for a tool that enables multi-tenant migrations that provide this level of flexibility as well as bidirectional data movement so that any shared data in a multi-tenant stack can be kept in sync.

  1. Can the data migration solution instantiate the target database and do automatic datatype mapping?

Migrating your data into a new cloud platform requires you to instantiate (i.e. create) the target database structures, including user accounts, tables, indexes, constraints, stored procedure structures and database objects and then load the data.

The right solution will simplify the data instantiation process by automating the creation of the target database structures, automatically mapping the datatypes and load these data into the new tables. The solution should also accelerate these processes by efficiently performing operations in parallel while compressing the data on the fly to squeeze every bit of performance out of your existing bandwidth to help accelerate the entire process.

  1. Can the solution handle the volume of changes?

Many organizations need a solution that can handle multi-terabyte databases with heavy volumes of changes occurring continuously. The solution should handle these change data volumes as well as the increased spikes of activity your application requires.

  1. Does the solution transfer data reliably?

Every organization wants the peace of mind of knowing that data is transferred reliably and that sources and targets remain in sync. To meet this objective, the solution should perform an initial bulk load of the data into tables on the target database. Simultaneously, it should capture all changes occurring on the source during the migration. When the initial load is complete, the system should automatically update the target with the changes. When all these operations are complete, the solution should compare the source with the target to validate that they are in sync.

Should something go wrong during the migration or the initial phases of use, the system should also be able to handle a failover of the source or target and perform data validation and repair if needed. In the worst-case scenario, the migration may need to failback to the original source without losing any data. To ensure that this failback occurs properly, the system needs to replicate data from the cloud to the on-premises system from the very first transaction. If it doesn’t, you can never migrate back to the old system because the data won’t be in sync.

  1. Is the migration secure?

Your data needs to be secure during the migration. Look for a migration solution that minimizes your data security footprint. A solution with one server and one port acting as a proxy server on your on-premises system that talks over an encrypted connection to only one server and one port (proxy server) on your cloud solution will keep everything locked down behind the firewall.

9. Does it do real-time and historical monitoring and alerting?

How do you know if your replication systems are up and running and operating properly? A migration solution that can track real-time stats and send alerts will automatically tell you if stats reach pre-set thresholds. For example, if latency goes past 5 minutes, the solution might send an email, SMTP, Slack, or SNS message to inform you in the format you’re accustomed to.

A solution that can integrate out-of-the-box or be extended to integrate with third-party systems such as ServiceNow, Atlassian, and others will streamline service ticket automation, allowing you to more quickly resolve any issues.

You also want to track current and historical performance statistics regarding your migration efforts, data throughput volumes, and latency, so that you can visualize and gain deeper insights into the nature of your changed data.

Learn how HVR can help you migrate your data quickly, easily, reliably and securely across your multi-cloud environment. Join us for a live demo.

[1] “How to Architect aMulticloud-Capable Hybrid Integration Platform,” by Massimo Pezzini, Gartner Group, May 3, 2018

About Joe

Joseph has nearly twenty years experience in the data integration and high availability market delighting thousands of customers by producing and delivering innovative real-time data integration products and solutions .

© 2018 HVR

Live Demo Contact Us