Data Migration for the Long Term

Data migration isn’t a surprise challenge to enterprises, especially with the ramped-up demand from regulators and for more data. However, there are a handful of questions businesses can address to set them on the right course over the long haul, according to a new framework offered by consultancy Osterman Research.

Data lost or damaged in a scheduled migration can cut into productivity, cause IT disruption and increase enterprise risk. This isn’t new, but the forces that drive it are. For example, email archiving in particular has become more “essential” in terms of regulatory and compliance demands, along with a growing body of internal and external content sources expected to be easily accessible for long stretches of time. Advanced analytics and big data are opening up new avenues of data intake and storage expectations. And mergers or acquisitions can test archiving and storage platforms.

To start, organizations need to ensure they have, as part of their archiving and storage platforms, a plan that includes “migration as a key and expected element,” says Michael Osterman, president of Osterman Research, based in Washington state. Osterman says that, even in changes to enterprise data demands, migration challenges generally fall within two catagories: volume and command.

“The sheer volume of content in many organizations makes migration difficult simply because it's not an easy to task to move many terabytes of data from one platform to another. It's a time-consuming and potentially error-prone process without the right migration tools. Plus, decisions have to be made about what data will be moved, whether or not a new archiving/content management tool will be used for new data while still supporting the older system for frozen data [for example],” Osterman says. “Second, chain-of-custody is an important consideration, since data that is in any way modified or altered during the migration may not be useable for e-discovery, regulatory compliance, etc.”

In the report, Osterman outlines five basic guidelines for long-term management of archived information that, while not comprehensive, aim at keeping organizations covered from risk and working with data during transitions.

1. Avoid Vendor Lock-in: There should be an entrance and exit strategy. One avenue toward this is to select vendors that offer an open-source archiving approach, though that can limit your choices. Preferably, Osterman suggests working with a vendor that offers “the ability to migrate cleanly and completely from any of several leading archiving systems to any other leading system, including cloud-based archiving.”

2. Work from a Single Interface: With a diverse range of data types, it’s more important to work from the same interface for managing archived content. More interfaces increases the number of silos and, with it, the training and storage costs. This also increases the chances of duplicating data.

3. Minimize Duplicate Data: Speaking of multiple sets of data, duplicate data poses obvious efficiency and accuracy risks in e-discovery, for example. Data deduplication should be an imperative in the archiving and storage process, as migration only increases the potential for inadvertently doubling – or outright losing – data.

4. Expect Migration: The new platform may be nice, but remember that there were systems and platforms before this one, too. Osterman writes: “[E]valuate any system for its potential to be replaced at some point down the road because it most likely will be replaced at some point. We recommend ensuring that tools, vendors and processes are known up-front that will assist in the migration if and when it is required.”

5. Keep a Cloud Option: Unlike many in-house systems, there are a multitude of data archiving capabilities for new sources or expectations, like messaging, social media and real-time communication. New systems must at least include the option of migrating to a cloud environment.

Osterman says he isn’t witnessing any particular industries handling these migration challenges better than others, with differing consequences from varying regulatory demand. But across business, Osterman says he expects “the situation to get worse in the coming years as data volumes continue to grow and as new content sources are archived – social media, SharePoint data – and must be available.”

To access the report, “Best Practices for Managing Information Archives,” click here.

This story originally appeared at Information Management.

For reprint and licensing requests for this article, click here.
Analytics Data and information management Core systems
MORE FROM DIGITAL INSURANCE