More in Store

New requirements change how IP video is archived

• By Barry Keepence
• Apr 02, 2008

Many organizations now require that CCTV video images are recorded and archived continuously from all cameras for 90 days or more. In large systems, this can create a significant storage requirement. The issue is compounded by users wanting to store video at the highest quality and maximum frame rate, particularly in homeland security and law enforcement applications where CCTV footage is required for evidential purposes.

The storage element of a large IPCCTV system can therefore dominate the overall cost of the system—both upfront and throughout its lifecycle—as hard drives will often need to be replaced. With potentially large amounts of data stored, the technology behind the compression and architecture of the storage solution is critical to ensure extended recording times. Similarly, the security of the stored data is important so individual disk and equipment failures do not result in data loss.

The Scale of the Problem
The factors that affect how much disk storage is required are video resolution, frame rate, number of cameras and the quality of the compression technology. So when comparing systems, the main variable is the compression technology, as all other factors are constant. The way in which video is compressed can make a significant difference to the storage requirements.

The following example demonstrates the storage requirements for a typical system, comparing equipment that has an average compression with the very best that is available. The difference between an average compression and the best is 180 terabytes of extra storage—and that’s just for a relatively small system of 100 cameras.

NVR Performance
In storage, CCTV systems differ from IT systems. The amount of data coming continuously from the cameras to the NVRs is huge. The amount of data going from the NVR to the users is low and periodic. The workload is constant—the rate of writing data to the disk is always high—not in bursts as with typical IT applications.

The processing overhead for writing and reading the video streams to disk is therefore important in NVR performance. There can be a considerable difference in this overhead between different vendors of NVR software. Software can ease the process and will be able to handle many more camera streams per NVR. The best NVR server software on the market has such a low CPU loading that 200 camera streams can be recorded on the lowest specification server PC.

Therefore, in a large system with a 90-day recording requirement, the limitation on the server is its storage, not its processing power.

Storage Architectures
The storage architecture for CCTV systems can be categorized as either centralized or distributed.

A typical example of a centralized system would be a casino, where numerous cameras are located in one building. The NVRs would be located in one central IT facility, with a central network switch. In this situation, all of the video could be recorded on one server; however, this would be storage-limited, not CPU-limited.

With a 100-camera system, one option would be to use 10 NVR servers, each fitted with an 18-terabyte disk array, assuming the system has the best compression technology available. This is still a lot of hardware, but because the CPU use is low, virtual servers can be deployed. Leading NVR vendors can run their software under applications such as VMware, allowing 10 virtual servers to be configured on one physical server and considerably reducing equipment costs. In reality, the typical casino uses between 500 and 1,000 cameras, so this approach is an important factor in lowering costs.

A typical example of a distributed system would be a rail network that could have 200 cameras located across 25 stations with, on average, eight cameras per station, with some stations only having four cameras. In this situation, what is needed is a small and flexible storage architecture that can use the right amount of storage for the right number of cameras.

For this example, assume 500 gigabytes per camera is required for 90-day recording. An NVR with 2 terabytes could be deployed at each station, with one NVR per four cameras. The smaller stations would have one NVR, and the larger ones would have five, for a total of 50 NVRs for all 200 cameras. This would significantly reduce the bandwidth requirement. The flexibility of the virtual matrix created by the IP video system means users with the right permissions could view and analyze recorded video from any NVR, no matter where on the network they are located.

If PC-based servers were used in the example above, then 50 separate servers would be required, each with local attached storage, which comes at a significant cost. An alternative would be to use dedicated, stand-alone NVR units that have the processing hardware and storage in a single compact unit. These are considerably less expensive than a PC server and equivalent storage.

To satisfy different types of applications, a vendor needs to be able to provide a flexible and scalable recording solution that can use both PC server configurations and dedicated NVR units as appropriate.

Data Security
With so much valuable data being recorded, it’s important to consider NVR security and reliability. Most PC-based NVR servers will deploy a redundant array of independent drive disk arrays. RAID is an umbrella term for computer data storage schemes that divide and/or replicate data among multiple hard drives. Different RAID levels give different levels of protection. In a RAID 5 configuration, for example, the data is stripped across three separate disks. If any disk fails, no data is lost and the computer can continue without interruption.

To take NVR security a level further, dedicated NVR hardware can be deployed. These stand-alone robust hardware units can have redundant power supplies and network connections, RAID configurations and hot-swappable drives. An IP video system that deploys standalone NVR units also should have a flexible NVR backup strategy. For example, if an NVR failure is detected by the system software, then recordings can be automatically switched to a backup NVR or distributed to other NVRs in the system. Like RAID configurations in PCs, NVRs can be mirrored, with the same video being recorded on two NVRs simultaneously, providing the highest level of security.

IP video systems based on H.264 can reduce the amount of storage required by between 25 and 50 percent, compared to MPEG-4 based systems.

DVR v. NVR
Storage issues with IP video systems and NVRs have been discussed, but what about traditional analog CCTV systems that use DVRs? It is important to differentiate between DVRs and NVRs, as both are often termed digital.

A DVR digitally compresses analog video feeds and stores them on a hard drive, with the term digital referring to the compression and storage technology, not the transmitted video images. The DVR therefore has to be located near the analog feeds and is only really used in centralized architectures. In contrast, an NVR stores digital images directly from the IP network and can be located anywhere on the network.

It is typical for a DVR to be located centrally, close to the analog matrix and control room equipment. As the size of the system and the number of recording days increases, there is no other option but to keep adding DVRs. This, in itself, is not a problem—other than the cost overhead associated with the DVR. A high-end DVR typically would be able to record 16 cameras onto a 2 terabyte disk. Using the example for a casino and assuming the best compression available, this would give a little over six days of recording for all 16 cameras, or 90 days for just one camera.

Clearly, DVR technology is simply too expensive for large-scale digital recording applications.

Up to Standard
There are a number of compression standards currently employed in IP video systems. H.264 is the latest official video compression standard, which follows the MPEG-2 and MPEG-4 video standards and offers improvements in both video quality and compression. The most significant benefit for IP video systems is the ability to deliver the same high-quality, low-latency digital video with savings of between 25 and 50 percent on bandwidth and storage requirements. By selecting a system based on H.264, further storage savings can be achieved. Even though H.264 is more efficient than MPEG-4, there are still differences between vendor implementation of the standard and the storage required.

There is no point in recording video from a camera at full frame rate if there is nothing in the scene to record. By using applications such as IndigoVision’s unique, actively controlled frame rate, the amount of video and storage can be reduced. When a scene is inactive, the video can be streamed at a much lower frame rate.

As soon as the motion analysis software detects movement, the video is streamed at full frame rate. Similarly, analytics such as virtual tripwire can detect an object crossing a line and raise an alarm. This alarm can start recording or increase the frame rate from that particular camera.

The requirement to record CCTV images for 90 or more days will have a large impact on the storage requirement and cost of the project. With larger systems, the inflexibility and costs involved with an analog/DVR solution mean IP video is the only way forward.