With 60 percent of all of the world’s data expected to be created and managed by enterprises by 2025 according to IDC, it’s not surprising to learn that enterprise players have moved swiftly to help their customers and organizations maximize the opportunity around this massive data growth. However, when it comes to one of the most significant drivers of this growth in information—surveillance data— many enterprises struggle to understand how they can apply it to do more for their business. In an age when data is fast becoming the lifeblood of the business world, these players are intuitively aware of the great potential surveillance data holds for them, but they lack an understanding of exactly how to tap into that potential.

Part of the reason for this is that the nature of video surveillance has changed in recent years. Once regarded as an isolated, security-specific function, surveillance today is more properly understood to be a source for maintaining business continuity. Traditionally, video has helped support safety and security measures, and has also provided a layer of accountability and insurance. Transformed by mobility, the cloud and big data, video surveillance data is booming, and can now serve as a strategic system for business outcomes that go beyond addressing only security needs.

Applications For Video Data

Video is proving to be a goldmine for improving operational efficiency and marketing opportunities, as well as creating new business value. For example, video data can help airports analyze flow patterns and wait times to deliver better services that create a more positive travel experience. Or, consider retail, where end-to-end video surveillance that captures data at all points of the retail experience—from supply chain processes to how store shelves are stocked—can help reduce waste, enable stores to better manage stock levels and help drive better sales.

To reap the full benefits of video surveillance data, a business must accurately plan and architect its storage systems to accommodate current and future growth, along with potential future uses for the data. At the heart of this planning are three fundamental questions to address: 

  • Where can I store digital video surveillance data?
  • How can I cost-effectively store it for an indeterminate amount of time?
  • How can I ensure the right people from across the company have access when they need it?

This shift from a one-size-fits-all, set-it-and-leave-it proposition to a model that accounts for data growth and subsequent requirements requires a radically different approach to video surveillance storage. The right storage architecture must be tuned to the specific business needs it is addressing, both today and in the future.

Video is a goldmine for improving operational efficiency and marketing opportunities, as well as creating new business value
Technology trends are reducing the costs of camera technology, making it possible for organizations to incorporate more high-quality cameras

Best Practices For Video Surveillance Data

The good news: Similar challenges have been tackled in other arenas, including traditional IT data centers. In these settings, data is the primary driver for how everything is architected, designed and operated. The primary goal is to store data effectively and efficiently while protecting it.

The same holds true now with video surveillance data. Best practices from the data center arena demonstrate how the right architectural planning can enable organizations to accommodate data growth and still access what they need, and when they need it. The right architecture can also help turn video into a value engine for new business opportunities, like helping ensure the efficient operational flow of people and products, and informing more direct and tailored marketing and sales efforts.

Several key steps can help achieve this.

Expect growth, and plan to accommodate it

Three factors are driving today’s video surveillance data growth: The number of cameras used by an enterprise and the type of cameras used, along with requirements pertaining to how long enterprises must retain data collected by those cameras. Specifically, technology trends are reducing the costs of camera technology, leveling the playing field and making it possible for organizations to incorporate more high-quality cameras into their surveillance systems. These higher-quality cameras, in turn, generate more data for enterprises to manage. This, coupled with the fact that retention requirements—largely driven by regulations, insurance needs and other legal implications—are shifting from weeks to years, means enterprises must balance today’s needs with tomorrow’s anticipated growth.

Enterprises can successfully achieve this balance by establishing a growth model for video data and choosing a storage system that meets their current needs and can easily scale to accommodate future growth. Key things to look for in systems meeting these requirements include technology architectures that enable an organization to start with the smallest possible system without sacrificing on features, and ones that are built to take advantage of hard drive innovations and allow for the seamless addition of incremental storage technology as needed.

Anticipate every situation—including when things go wrong

The shifting use of video surveillance and its ability to maintain business continuity means video surveillance systems must remain highly available to record and capture data. However, system growth increases the number of moving parts and the potential for associated devices, like hard drives, to fail. Enterprises can minimize risk by treating their video data as the final measure of security, and planning everything around it. Making data the focus can ensure an enterprise can continue to record and access surveillance data, even when components fail.

Organizations should seek out systems designed to not only anticipate failures, but also to quickly recover from them if they occur, and include features like proactive failure detection, fast failure recovery and preemptive fixes to assure data availability. Systems designed with data as the central focus can handle component failures without video loss and are also able to recover fast from one failure before a second one can occur.

Plan for the longevity of data

Regardless of retention requirements associated with video surveillance policies, enterprises should account for the possibility that they must retain at least some video data for years. To prepare, an organization should adopt a multi-tiered storage architecture so not all data is stored in one highly-available, high-performance, cost-intensive tier, as not all video needs to always “travel first class.” A hierarchical storage system that moves older video data to a lower-cost storage tier is ideal for handling video archiving.

Adopt An Open Architecture

As technology continues to evolve, enterprises should adopt an open architecture featuring a modular design. This will enable businesses to easily update individual parts in line with future technology trends without significantly disrupting the system’s overall architecture. For example, hard drive technology advances increase capacity and drive down the cost of drives. Changes in cameras and VMS software may increase bandwidth needs of storage. A storage system must take advantage of these advances without requiring wholesale hardware replacement every time things change and evolve.

We have already seen video surveillance evolve from an isolated application to one that can contribute to an organization’s bottom line with the right architectural support and planning. What the future holds beyond this evolution is anyone’s guess, but organizations can set themselves up for continued benefits by adopting a storage system that supports the open and seamless interface for data access needs of future applications, such as artificial intelligence and predictive analytics. Ultimately, having a plan for all data uses—including the non-traditional surveillance uses—helps ensure an enterprise is able to squeeze as much value out of its data as possible.