Around the Storage Block

Predictive flash storage puts an end to after-hours firefighting

Proactive resolution with HPE InfoSight means no more painful and costly escalations—just the level of IT support your business needs. 

You've had a long week, and by 5:30 p.m. FridaySkyword_Predictive_Flash_Storage.jpg you're happy to be on your way home to enjoy a well-deserved break—but then your phone lights up and your boss's number flashes on the screen. It's yet another Friday night fire drill, where something has gone wrong with your data center infrastructure and you're being pulled back in. To get your lives back, you and your boss need to search for a storage solution with predictive analytics that can think for itself to predict and prevent these problems.

Tame your infrastructure with predictive analytics

Preparing for this conversation will likely first mean quantifying the cost of recent downtime and showing how that affects your company's operational expenses. But your experience has shown that finances aren't everything, and there are several things that you want to keep top of mind when you're reviewing and validating solutions. Leveraging state-of-the-art machine learning for infrastructure administration will make your business more productive and free up your time to focus on value-add projects for better business operations—rather than trouble tickets and constant firefighting.

If your company has been focusing on reactive monitoring, which does little more than flag a problem and leave you to discover the resolution, it's time to look at predictive analytics, which discovers potential issues—from slow performance to system downtime—before they ever occur. When your infrastructure cannot automatically identify and prevent an adverse situation, prescriptive resolution will determine a clear path to problem correction. This eliminates the need for you to research documentation, go digging in online forums, or call support. Plus, in the unlikely event that there's no available automatic prescription, rapid root-cause analysis should quickly ID the cause of the problem—without lengthy troubleshooting that could involve cycles of problem recreation, log review, and engineering analysis. Without predictive analytic capabilities, problems like these could take you weeks or even months to resolve.

To help you understand the big picture of your IT ecosystem, predictive analytics should support either the collection of data across the infrastructure or the cross-stack application of analytics. Without this overall inspection, you could experience multiple false alerts or missed-issue indicators that could impact your ability to respond and rapidly resolve problems.

Make flash storage predictive with artificial intelligence

When you do have a disruption issue during one of those fire drills, the many layers in your infrastructure stack can make it difficult to pinpoint the root of the problem. Are you facing host, VM, or networking problems? Are interoperability or configuration gremlins wreaking havoc? Has a disk gone down? Adding to your detective legwork, you'll probably need to call on the support team from at least one vendor, and deal with the finger-pointing and escalation process that doesn't ever seem to get the problem resolved in a timely manner.

All of these issues can slow the delivery of data to applications, creating a productivity bottleneck that HPE calls the "app-data gap."

That's where HPE InfoSight, advanced AI for the data center, can provide a solid foundation for an autonomous data center. By continually collecting data from thousands of sensors embedded in every system, this solution is able to predict and prevent data disruptions, whether they're related to storage, configuration, interoperability, or other issues. This enables you to improve performance, dodge issues, and optimize your computing resources on an ongoing basis.

The technology's advanced machine learning is based on more than a decade of analyzing and correlating telemetry data from the HPE Nimble Storage and HPE 3PAR StoreServ installed base. This means InfoSight is in a continuous learning mode, making every one of your systems smarter and more reliable. By staying ahead of problems and allowing you to quickly correct the few that do happen, InfoSight can help prevent other customers from dealing with the same issues. And when potential issues and abnormal behavior are rooted out, storage self-improves, while steps can be taken to continue enhancing a data center's performance and uptime.

When you decide to leverage predictive flash storage, routine management will become a thing of the past. And with management and maintenance carrying the highest price tag when it comes to running your data center, these results matter. HPE InfoSight's unique, cloud-based AI platform can put an end to your after-hours troubleshooting. By eliminating wasted time, infringed-upon nights and weekends, and the ensuing headaches, you can transform how you manage your infrastructure. Nimble Storage and 3PAR are powered by HPE InfoSight, so problems are predicted and prevented before they can impact your business, as well as your personal life. Therefore, hope is not lost. Next Friday can be a smooth sail into your rejuvenating weekend.

Meet Around the Storage Block blogger Kate Stanton Whalen, IT Journalist.

StantonWhalen.jpgKate was introduced to the world of IT at Digital Equipment Corporation (DEC) in the early 1980s, when DEC was a trailblazer in scientific, modular computing. There, she learned to merge her creative nature and love of the English language with the discipline of business communications. Her experiences at DEC, Compaq, and HP/HPE, as well as four New England-based communications and marketing agencies, enabled her to build a broad skillset in a variety of marketing and communications fields including event management, video production, brand stewardship, and social media marketing. 

0 Kudos
About the Author


Our team of Hewlett Packard Enterprise storage experts helps you to dive deep into relevant infrastructure topics.