By Eric Burgener – Research Director, Storage, IDC
High data growth, dynamic business requirements, and an increasing demand for real-time response have changed the face of storage management over just the last five years. Many small- and medium-sized enterprises are facing the challenge of managing hundreds of terabytes in the next year or two. Many large enterprises are already well into the petabyte range. New technologies, including virtualization, flash storage, and cloud, have become mainstream in IT infrastructures that must support rapid provisioning, easy expansion and reliable, predictable service levels.
As part of this evolution, administrative and support models for storage are also changing significantly. A major trend that IDC has noted over the last five years is that storage management tasks are migrating away from storage specialists to IT generalists. At the same time, administrators are being asked to navigate more complex IT environments with significantly larger data sets and more stringent real-time response requirements. These changes pose risks for companies that must meet rigid service level agreements for mission-critical application environments.
The flash-optimized storage infrastructure necessary to meet these requirements must not only be extremely fast and reliable, but easier to manage at scale. To deal with these challenges in the face of these trends, vendors have turned to automation for workflows, routine administrative tasks and support services. Several enterprise array vendors – primarily startups– have introduced cloud-based telemetrics that leverage big data concepts with their next-generation storage solutions that are providing compelling new capabilities particularly in the support arena, providing significant value to both customers and the vendors themselves.
Big data concepts are bringing huge value to businesses of all kinds, using correlative techniques against extremely large data sets to meet objectives of all kinds. One of the things that is most interesting about big data/analytics is that it can uncover relationships that an analyst may never even know to look for. Knowledge about those relationships can be used to improve service and make better business decisions.
These techniques are very applicable in the support automation arena as forward-thinking vendors use that data to inform a variety of services to drive predictive analytics, performance, capacity planning, and rapid diagnostics to improve system performance, availability, and reliability. The approach drives bottom-line business results in the areas of meeting service level agreements, cost-effectively supporting business growth, resolving issues far more quickly to maximize overall availability, improving administrative productivity and spans of control, and delivering the type of customer experience that encourages high re-purchase rates.
Customers interested in leveraging this new approach should look for several features: the proactive use of predictive analytics, a “software as a service” (SaaS) design, and comprehensive data capture. The proactive use of predictive analytics can help resolve potential issues before a customer is even aware of them, reducing problem resolution time, pre-validating planned upgrades with existing configurations, and improving customer service overall. And analytics can provide valuable input into the performance and capacity planning process (not to mention a vendor’s own R&D process).
The SaaS design provides easy accessibility by any type of device from anywhere for administrators, allows the analytics processing to be offloaded to the cloud, and makes feature enhancements and upgrades to the remote diagnostics layer much easier and less disruptive for both the customer and the vendor. Look at the extent of data capture – those vendors that are collecting data points about not only the configuration of their array but also virtual machine and application configuration have more data to leverage to improve reliability and the overall customer experience. Ultimately, an ability to extend correlations and predictive analytics beyond just the storage system results in faster, more comprehensive problem identification, diagnosis and resolution, supporting higher up-time statistics.
Storage vendors that have implemented these types of support automation capabilities are leading the industry in the right direction. It is fair to say that, when it comes to managing storage infrastructure – and quite likely ultimately IT infrastructure as a whole – this type of cloud analytics-based approach is the wave of the future.