Guest Article by Phil Goodwin of IDC
IDC estimates that as many applications will be deployed in the next five years as have been deployed in the previous four decades. In many cases, these new apps modernize existing apps, but may also be brand-new apps, both in support of digital transformation (DX) initiatives. We estimate that 55% of organizations will have implemented some DX initiatives by 2020. DX is an enterprise-wide endeavor involving every aspect of the organization with the intent, among other things, of enhancing market competitiveness through faster decision making based on more complete information.
To deploy these new apps and fully achieve the potential of DX, development teams cannot rely on old tools and development methodologies. While many developers use leading-edge development languages and tools, others continue to rely on data management products that have become the bottleneck to development efficiency. Specifically, test/dev teams often use data replication tools designed for the client/server era that make physical copies of databases and file systems for testing. In many cases, accurate testing requires the use of current data. So, test/dev teams need to constantly update the test data systems. They want daily or on-demand data refreshes, and any delay slows the development and deployment process.
According to IDC research, 72% of IT organizations take a half-day or more to create or refresh a single data copy. This long processing delay means most organizations only refresh data copies once per week, usually on a weekend to avoid impacting production applications. But that's not the only problem. While the provisioning time needed to create a new copy averages two days, it takes much longer in some organizations. DBAs and developers would prefer on-demand self-service copies and they want the copies to be the real data, not a test subset. Unfortunately, neither instant availability nor self-service is realistically possible with traditional copy tools and processes.
Copy data management (CDM) products are a relatively new class of tools that can significantly improve the time and effort associated with copy operations. Rather than make a physical copy of the data and require a physical refresh, CDM tools make a virtual (metadata) copy. Because it is a virtual copy, there is no separate storage provisioning or consumption required and no impact to production applications. Virtual copies can be made in seconds or minutes and refreshed in real time. Suddenly, self-service, on-demand data copies can be a reality. Data cycle times can be reduced from a week to a few minutes. This means faster development, faster deployment, and faster modernization.
Testers tend to leverage on-demand test environments. Public cloud is a perfect choice for testing. It’s important to consider a CDM platform that can instantly provision virtual copies in a self-service manner in any private/public cloud using any storage, even object storage. This enables enterprises to truly leverage a cloud- and storage-agnostic CDM solution.
We can’t forget about security, however. Businesses often don’t allow self-service copies to be made because most organizations require manual data masking after a copy is created. Data masking assures that sensitive or regulated data cannot be seen by unauthorized users. Our research shows that 80% of organizations have either manual data masking or no masking at all. CDM tools often can integrate with or provide role-based masking and/or scripts so that data is masked upon copy creation based on the credentials of the creator.
All other things being equal, organizations that use modern CDM applications get applications into use days, weeks, or even months faster than those that do not. Organizations planning to deploy dozens or hundreds of new applications and modernize existing ones should consider using modern CDM. These tools can speed development cycle time and yield genuine competitive benefits.
IDC Report – Actifio Strengthens Its Data-as-a-Service Vision with AppCentric Approach in the Multicloud and DevOps Era
Sign up for blog updates via email.Subscribe