One of the most important (and often overlooked) aspects of any successful surveillance system is choosing the most appropriate type of storage.
"The truth is there will never be a "one size fits all" answer to storage, simply because it depends on the specific security, operational and analytical requirements in play. It’s why keeping up-to-speed with the latest options is so important."David Aindow, Technology Director, Synectics
We are in an age of 'Big Data’, a term which relates both to the volume of information that modern solutions are required to receive and manage, and the content itself. For example, integrating to third party systems is no longer an aspiration, it is a necessity; data from security, operational and emergency systems for an entire site (or multiple sites) is now often collated, monitored, and controlled through a single surveillance management platform.
Add to this the introduction of HD IP cameras – and more recently the arrival of 4K camera technology – and storage choice becomes a topic you just can’t ignore. All of that data has got to be kept somewhere.
The problem is, if not planned for properly, inadequate storage can leave a system unfit for purpose and can ultimately restrict users from reaping the benefits of new technologies.
In this interview, we find out about the key considerations behind surveillance data storage, from Synectics' Product and Technology Director, David Aindow, and IT Product Manager Sean Kettlewood.
What current trend has had the biggest impact on surveillance data storage?
SK: Without a doubt that would be the transition from analogue to IP with HD and beyond. The shift to digital, using megapixel cameras, makes higher resolution image capture at faster frame rates possible, and that puts increased pressure on storage system infrastructure.
Historically, the Digital Video Recorder (DVR) was the obvious (and possibly only) choice for video storage. But its limitations in terms of finite source inputs, frame rate and typically fixed storage capacity, make it unsuitable for many of today’s evolving IP needs – particularly in terms of protecting against failure and managing scalability. Previously, adding a new DVR might have been a relatively simple task but multiple devices did not easily come together to create a single unified system. Instead they performed as separate devices independently of one another.
DA: NVRs (Network Video Recorder) with either internal storage or DAS (Direct Attached Storage) devices have become more prevalent as they can offer greater data throughput, and don’t have the physical (geographic proximity) restrictions of DVRs. They typically afford much greater levels of scalability, significantly higher resilience possibilities, and perform as a single, holistic system, regardless of the quantity of devices deployed.
With the digital surveillance revolution and the trend for whole-site systems integration - unifying video, audio and operational data inputs into one platform - new storage technologies have entered the market and deserve consideration.
Users now have the choice not only between DAS, but also Network Attached Storage (NAS) and Storage Area Network (SAN).
What’s the difference between DAS, NAS and SAN?
DA: Put simply, DAS and NAS systems attach file-based storage either directly to the host device (NVR) or to an existing network. SAN offers increased block storage capacity through a separate network which can be accessed via a server. All enable storage to be shared or scaled up as required.
At the moment DAS and NAS are more commonplace solutions for storing surveillance related data. There are a number of reasons for this, but the biggest one has been the need for high transmission speed from the host device (NVR) to disk. Until recently, SAN systems just couldn’t cope with the throughput demands of real-time video storage. In the past 18 months or so that’s changed, and most SAN systems can now transfer data between servers and disks at the similar peripheral channel speeds as if directly attached.
Whereas DAS and NAS devices can be added to a network with minimal impact on system infrastructure (no potential data loss, need for system downtime or extensive reconfiguration), SAN implementation is relatively more demanding in terms of updating infrastructure and personnel training – it’s a solution far more familiar to IT departments than surveillance personnel. It is highly unlikely you would deploy SAN purely for video data storage. It lends itself to those applications where there are multiple ‘data stakeholders’ all with storage requirements – such as an airport where operations, surveillance, risk management, and other systems all have vast amounts of data to manage.
SAN’s strength is in data depth. Fiber channel technology typically allows data transfers at 8Gb/s to accommodate enterprise volumes of data to be stored with high levels of redundancy. It’s why we are increasingly seeing it used in large-scale applications such as airports, oil & gas plant monitoring, or other forms of critical infrastructure protection.
"Purchasing decisions need to be made now in order to make the right long term storage investments. For example, an organization doesn’t want to be in a situation where it is introducing the latest HD cameras only to find that it has to record and store footage at a lower resolution."
If there are a number of storage options available, why is planning ahead so important? Can’t organizations just adapt what they have as they grow?
SK: The need to plan storage is more to do with what an organization wants to achieve with its surveillance system (and by when) than it is about the storage solutions themselves.
In the next five years, how many cameras will I need? Do I want these to be all IP/HD (even 4K) or will a hybrid system still fulfill my needs? Will I have more sites in different geographic locations to cover? What other systems on my sites do I want surveillance to integrate with and how much data will this involve?
The answers to all these questions will dictate purchasing decisions that need to be made now in order to make the right long term storage investments. For example, an organization doesn’t want to be in a situation where it is introducing the latest HD cameras only to find that it has to record and store footage at a lower resolution, nor does it want to be investing money in network infrastructure now if it is going to have to be replaced entirely in a few years to meet evolving needs.
What storage developments are we likely to see in the future?
SK: While DAS and NAS applications are likely to remain the common choices for surveillance only storage – along with an increase in the adoption of SAN for large scale integrated data solutions – more interest in cloud-based surveillance data storage is inevitable. Naturally, concerns regarding data security will inevitably require consideration.
DA: The truth is there will never be a "one size fits all" answer to storage, simply because it depends on the specific security, operational and analytical requirements in play. It’s why keeping up-to-speed with the latest options is so important.