Some technology challenges loom large for a period of time, and then get resolved and fade away. Internet access, for example, and mobile payments. Other problems hang around for years, sometimes even longer. The one that has irked me in my own career for over five years now is data mobility.
Since the earliest days of the cloud, one of the major questions about cloud adoption has been, “How do I get my data out of my data center and into the cloud? And what if I want to bring it back or move it somewhere else?” As Laz and I often discuss, data has gravity and when you’re talking about tens of terabytes for a workload – or worse, petabytes – the problems with moving data around appear to have no quick fix. Even my last company, which provided an “easy button” to the cloud, was limited by an enterprise’s bandwidth and network configurations.
Fast forward a few years and the options are not much better. Frequently, a move to the cloud involves a professional services project, in which migration of data consumes a large amount of time and effort. This is hardly the most “agile” introduction to the cloud! Other options are equally unappealing, such as adding expensive, larger-bandwidth connectivity to the cloud or shipping data loaded on physical devices to data centers that are more directly connected to the cloud. Worse, companies may also be accepting the idea that large and growing data sets will just need to stay where they are, creating “islands” of resources that are locked in different environments, including on-prem and in multiple cloud services. None of these solutions make cloud adoption easier or more palatable for enterprise IT teams already concerned about loss of control and security.
Meanwhile, companies are increasingly looking to consume solutions and infrastructure on-demand, and mobility solutions need to address these emerging requirements. When applications can be anywhere – on-premises, in private and public clouds – data becomes a boat anchor that holds companies back. Latency emerges as a major factor if compute and storage are separated by more than a few milliseconds. Most enterprises are in an evolution that requires a hybrid cloud approach; some applications and data remain on-prem, while others have moved or are built from scratch in the cloud. New service providers like ClearSky are taking the data mobility challenge head-on, managing data movement securely and in a highly optimized way between on-premises and cloud environments without requiring IT to manage the process or conduct a professional services engagement to get it done.
Here’s what we believe is “state of the art” for data mobility in a hybrid world. First, simplicity is key; the enterprise should be able to “click to move” the workload and expect it to happen automatically. Second, data optimization should be built in; compression, dedupe, WAN optimization and intelligent caching solutions that determine which bits of data truly need to be moved are all required to minimize the size of data transfer. Third, a dedicated, single hop, high speed network is required in order to provide the low latency and high throughput the task requires. And finally, any data movement requires a comprehensive security approach, with full encryption and key management as must-have features.
So if you’re an enterprise adopting the hybrid model and moving to the cloud, don’t let data mobility be a roadblock. Know your options and expect your service providers to take the lead on addressing this issue for you. Your data needs to be as mobile and agile as your business, and it should be available to you wherever you choose to run your applications.
Learn how five companies gained on-demand storage access without sacrificing performance, availability or security.