By now, there have been enough high-profile outages of vital consumer and business-class data environments that “disaster recovery” has entered into the lexicon of everyday users.
On the professional end, however, many recovery specialists are starting to couch DR as a mere subset of a much larger operational goal: data and infrastructure resilience.
A new study by Vision Solutions lays out what it calls the State of Resilience in the enterprise industry. The company canvassed more than 500 IT professionals on their strategies to not only restore data environments when they crash, but to prevent crashes in the first place. Nearly half say they are pursuing hybrid approaches that encompass storage, software and cloud infrastructure to protect critical assets. At the same time, more organizations are turning toward “agile data protection technologies” — things like logical replication and clustering — in conjunction with traditional tape-based storage to maintain continuity.
Indeed, the idea of resiliency is quickly moving from buzzword to a deployable strategy, according to Mark Thiele of Las Vegas data center provider Switch. With IT becoming so complex and distributed, a single recovery or continuity program simply cannot prepare for all potential risk factors. In the future, data environments will be characterized not just by how many users they can accommodate and how they can streamline physical infrastructure, but by how well they perform self-analysis and self-healing tasks, preferably without impairing performance of critical apps.
One of the ways to accomplish greater resilience is to make the data center less complex, according to Jason McGee of IBM PureSystems. Through centralized management and operations, enterprises will be better able to adopt new technologies and processes more quickly, even as infrastructure itself becomes more distributed. By streamlining the architectural underpinnings of the data environment, enterprises will find it easier to protect against external threats and to restore operations quickly following an outage.
On some level, of course, resiliency must take into consideration the health and well-being of physical infrastructure. And on that front, there is a growing variety of solutions aimed at ensuring hardware remains connected and operational. In HVAC circles, for example, data center requirements are driving development of high-density, medium-voltage systems optimized for air floor containment and other specialized cooling environments. As well, continuous power to air handlers and other devices is becoming more crucial as power densities top 200 watts per square foot, enough heat to shut down critical systems within a minute if not properly cooled.
Resiliency, then, encompasses a number of attributes the go beyond the scope of mere recovery. Strength and flexibility of data infrastructure are certainly major components, but so are visibility into operating environments, management of critical systems and, yes, the ability to either recover failed environments or shift loads to alternate ones should the worst occur.
Data is simply too important to entrust to recovery alone, however. Modern architectures will need a much broader platform in order to keep information safe and available.