- stock.adobe.


Intelligent storage systems optimize performance, proactivity

Intelligent data storage use is on the rise, employing AI, machine learning and predictive analytics to automate storage and prevent problems before they start.

Enterprise applications rely on storage that can handle vast amounts of heterogenous data and support complex, fluctuating and often unpredictable workloads. Storage must also be able to ensure the data's security and integrity, while addressing issues around privacy and compliance.

In enterprise-level operations, adding capacity or deploying faster drives is not enough. Enterprises need storage as smart as the applications that use it, and intelligent storage systems fit the bill.

Vendors are now starting to offer intelligent storage that incorporates AI and machine learning to proactively identify and address potential issues and optimize operations across the storage infrastructure.

New storage challenges require smarter fixes

More applications are putting greater demands on storage systems to support compute-intensive workloads such as AI and big data analytics.

Concurrent with these challenges comes an escalation of sophisticated security threats, such as ransomware attacks, an assortment of new compliance and privacy regulations that vary from region to region, including the European Union's General Data Protection Regulation as well as new and evolving data center architectures.

To meet these challenges, storage must be able to quickly assess environments and automatically take actions to remedy problems, while providing extensive real-time visibility into all layers of the storage stack.

An intelligent storage system collects and analyzes infrastructure metric data and uses the results to reduce downtimes, maximize resource use, optimize application workloads and ensure availability. Intelligent storage also provides monitoring and alerting capabilities for tracking possible security and compliance issues.

The technology behind intelligent storage systems

Intelligent storage collects massive amounts of telemetry data from storage systems and related systems and services. The data can include performance metrics such as throughputs, latencies and per-volume IOPS, as well as details such as cache use, bandwidth use, workload measures or other related data. HPE InfoSight, for example, collects data from millions of sensors per second across its installed base of global customers.

But collecting data is not enough. Intelligent storage continuously analyzes the data to identify patterns that can be used to forecast trends, isolate bottlenecks, track hardware failures and predict potential issues before they occur.

At the heart of this learning process is a sophisticated analysis and recommendation engine that uses AI, machine learning, deep learning and predictive analytics.

AI includes two important subdisciplines that intelligent storage commonly uses: machine learning and deep learning.

AI includes two important subdisciplines that intelligent storage systems commonly use: machine learning and deep learning. Machine learning makes it possible for computers to learn from data without requiring a specific set of programmed instructions.

As more data becomes available, the machine learning system continues to learn and, in the process, becomes better at making predictions. Deep learning is a subset of machine learning that employs multilayered artificial neural networks to emulate human learning to infer meaning from examples.

Intelligent storage also employs predictive analytics for many of its core operations to uncover patterns in data and forecast specific outcomes.

To carry out these operations, predictive analytics uses AI, data mining, analytical queries and predictive modeling. For example, predictive analytics often uses machine learning to train predictive algorithms, using samples from the collected telemetric data.

The intelligent storage difference

An intelligent system system can simplify storage management and better protect stored data. It can automatically predict and prevent performance and availability issues, respond to changing workloads, assist in troubleshooting, help plan capacity and alert users to security or compliance issues.

The result is improved performance, higher availability, reduced costs, increased security and more time that administrators can spend on other tasks.

How vendors deliver intelligent storage can vary from one offering to the next, as can their definition of what makes storage intelligent. Just because a vendor uses the word "intelligent" does not make it an intelligent system.

An intelligent storage system provides a smart approach to storage that goes beyond basic operations to deliver a proactive, automated system that optimizes performance, adapts to changing workloads and maximizes data security and compliance. Anything less is just another storage product.

Dig Deeper on Storage management and analytics

Disaster Recovery
Data Backup
Data Center
and ESG