Role of storage vital in unlocking the business value of surveillance data

Sept. 16, 2015
Improved image quality, enhanced analytics demand improved video storage infrastructure

For years, video surveillance has been instrumental in helping keep people safe and deterring theft. But the role of video surveillance is quickly evolving beyond passive observation and loss prevention. With IP-based cameras capable of streaming sharp, high-resolution images and improved video analytics tools, surveillance is getting smarter and, with it, a wider range of business units are now turning to surveillance data for business intelligence. While the possibilities for using surveillance data are exciting, IT departments should first consider how they will manage storage requirements for that data.

In retail for example, understanding consumer behavior is paramount to driving increased sales. The better a company understands its potential buyers, the more satisfying shopping experience it can provide for them. And, information such as shopping traffic volumes, shopper movement through the stores, and the effectiveness of promotional displays at gaining shopper attention can now be captured from surveillance footage instead of “mystery shoppers” walking around the stores. With this information and input from point-of-sale systems and other devices, retailers can make better decisions regarding store layouts, advertising displays, and targeted marketing promotions to increase sales and customer satisfaction.

 In the transportation and shipping industry, optimizing the movement of cargo into and out of ports is essential. By using surveillance data and integrating it with RFID tag information on containers, companies can analyze traffic patterns and better organize shipyards to improve the flow of cargo and lower operational costs. In a similar fashion, surveillance data combined with input from RFID and other sensor technologies can be used to track materials through manufacturing facilities to reduce costs and eliminate waste.

There are other examples out there. But one thing is clear: with the increased volume and granular detail of surveillance footage available, what traditionally has been thought of as “surveillance data” to be used for loss prevention is quickly moving in the direction of what can be called “video-based data” and mined for other business uses. Many providers of leading analytics solutions already have recognized the potential for video-based data, shifting from strictly offering security analytics, such as event detection and alerts, to including more business-based analytics, such as traffic volumes and movement trends. In fact, according to IHS, the global video analytics and intelligent video surveillance market is forecasted to grow at a compound annual growth rate of 14 percent, to $22 billion by 2020.

Deriving business value from surveillance data is not simple. Most organizations lack the data storage infrastructure required to support active intelligence gathering and data sharing. If video surveillance is to play a growing role in enterprise business intelligence, there will need to be a transformative change in how this data is managed.

Balancing Performance, Cost, and Accessibility is Vital

Trends and patterns emerge from analyzing data over time; therefore, the longer data is retained, the better the quality of the analytics and more effective the derived business decisions will be. This principle applies to surveillance data. But, too often, storage infrastructure has been an afterthought in the video surveillance buying process, resulting in trade-offs being made between retention times, performance, and accessibility to manage cost. As a result, many organizations lack a storage infrastructure capable of supporting the use of surveillance data for enterprise business intelligence.

For a surveillance system to be effective, the storage infrastructure must be able to handle multiple streams of high bandwidth data in parallel without dropping any data. The more high definition cameras installed, the more bandwidth and processing power required to keep up. To support complex analytics, the storage infrastructure must not only be capable of fast ingest but also able to access data files quickly, process them, and provide information back as soon as possible so it can be acted upon. But while performance and access are important, so is managing cost.

Long-term retention of data to support analytics must be balanced against the cost. For example, implementing a monolithic, disk-based storage infrastructure will deliver very good performance, but it’s also very expensive and can result in islands of storage, making long-term retention of data cost prohibitive. And, while offloading older data files to less expensive mediums such as tape storage can lower costs, it can make keeping track of the data difficult and the process of performing analytics very cumbersome.

The best approach to managing video data so that files are retained cost-effectively—but can be quickly accessed and retrieved for analysis later—is to implement a high-performance, tiered-storage infrastructure that can be managed as a single system. Using a tiered architecture enables video files to be stored on the most cost-effective medium—whether that is primary disk, secondary disk, tape, object storage or a cloud instance—based on user-defined policies. Once the policies are defined, the system manages the movement of data between the tiers and the metadata remains intact. As a result, video files are able to be kept for longer periods of time and can be easily searched and accessed through a single file system view, lowering the cost of storage and enabling analytics to be performed more effectively.

Tremendous business value can be extracted from surveillance data through business analytics. But to do it effectively requires a cost-effective storage infrastructure that allows data to be retained for a long period of time while still ensuring fast and easy access to both the data files and their metadata. As a result, storage infrastructure must be carefully considered during the evaluation stage of any surveillance solution and must be designed with performance, accessibility and cost in mind.

About the Author: Wayne Arvidson is Quantum’s vice president of video surveillance solutions. A seasoned global marketing, product management, and business development executive, Wayne has 25 years senior management experience in companies ranging from start-ups to Fortune 500 firms and drives Quantum’s strategy in the surveillance and security market.