The Migration Cometh Before the Cloud

The cloud can increase your data flexibility, improve productivity and save money in the bargain — but first you have to get there.

It is funny, although not surprising, that most cloud providers stress the myriad efficiencies and operational benefits of their services, but are not so upfront about the migration. But as with any other migration, however, implementing an entirely new data environment, even a virtual one, involves a long list of tasks — from network pathway redefinition to client-side OS updates.

Whenever there is a business need, however, there are tools and systems to help solve or at least mitigate them. For cloud migrations, one of the latest is Dimension Data’s new cloud readiness program, which seeks to assess initial preparedness for the cloud and circumvent potential problems before they arise. The program zeroes in on four critical elements of the migration — business alignment, organization, infrastructure and applications — and then uses a mix of services, training and analysis to determine the appropriate steps and ensure that all stake-holders are on the same page. In the end, the goal is not just to implement a smooth transition, but to ensure that the selected cloud services remain beneficial over the long-term.

Of course, if you are a traditional data platform provider with a loyal following on a proprietary system, migration to the cloud is that much easier. This is what IBM has discovered as it seeks to upgrade system z and p mainframe users to the cloud. The company’s SmartCloud Enterprise+ program utilizes IaaS services from eight data centers around the world to run AIX and even SAP workloads for customers who need the increased scale. As well, the program offers hosted management, security and other services, as well as varying degrees of service and availability depending on the nature of the workload.

While many enterprises are careful to sign detailed service-level agreements (SLAs) with their cloud provider, many do not take the added step of detailing the cost and expectations of the initial migration, says tech consultant Matthew Ramsey. What’s needed is a Statement of Work agreement that covers objectives, tasks and deliverables for the migration, with both schedules and costs laid out in explicit detail. The last thing both the CIO and the CFO need to hear is that the cloud project they authorized is behind schedule and over budget before actual service has even started.

Then again, sometimes upper management can be the problem, rather than the solution. This cautionary tale on InfoWorld highlights what can go wrong when visions of cloud nirvana obscure the realities of data infrastructure. In short, proper planning and a working knowledge of existing application and physical-layer environments are crucial for a successful cloud deployment — not to mention a trusting, supportive relationship between management and the technical staff.

It would be nice to say that any of these approaches — from proper planning to State of Work agreements to advanced migration services and programs — will make the actual migration a joy to undertake, but they won’t. At best, they can help you avoid some of the pitfalls that turn a normal migration into a nightmare.

Ultimately, expanding data infrastructure is a complicated task that requires smooth interaction between a number of moving parts. And in that light, it’s important to ensure that the new operating environment, cloud-based or otherwise, will deliver enough of an improvement over existing architectures to justify the cost and the hassle of implementation.