"If you can't measure it, you can't manage it"
Jack Welch, CEO at General Electric
There are three Key Performance Indicators are necessary to properly measure the health of Video Surveillance applications and the supporting infrastructure. These measures are as follows:
The measures names for VPI and VSDI were changed since the provisional filing to VPU and VSDI, respectively, to make these acronyms clearer to end-users what they were describing.
These three measurements are orthogonal to each other. One analogy we like to use is an aqueduct: if the water is the data flowing through the aqueduct (VSDI), then water takes a path through the infrastructure (VPU), to end at a destination where the the water can be stored and used (VRC).
Video Path Uptime is stream uptime aggregated at either the server, site, or company. It’s not server uptime. If your server indicates a VPU% that is lower than 100%, that means that one or more of your streams are not up and you are not recording that stream.
Some of the reason why VPU% may drop are the following:
- Camera device is not responding
- Network connection is severed
- Server is down
- No network sessions between VMS and camera
- Storage media is not available
- VMS recording service is not running
Any video surveillance system has a physical capacity which constraints the amount of data that can be retained. These constraint were usually carefully specified during the design phase, but can quickly move off target as these sophisticated systems get older. VRC is a retention measure for you video surveillance system that allows you to gain visibility into the overall retention compliance of your system.
VRC is based on individual stream retention aggregated at the server, site, or company. It is defined as the actual/ desired measurement of retention. When you first installed a video surveillance system, you set certain camera streams to be recording at certain parameters such as resolution, frame rate, codec, etc, and importantly days of retention.
A drop below 100% in VRC could be caused by only one thing: one or more streams isn't achieving the retention that it's designed to achieve.
Note: In the event you have set your recording to just record over the oldest file in the file system, you have the ability to calculate a retention even if it's not set in your VMS. This could become handy if you do have a minimum number of days that you absolutely need the data for. If you set the retention goal in Viakoo, then you can be alerted if in the event that you drop below a certain amount of day.
VSDI measures the performance impact of saturation or decay of a video network on video quality. VSDI measures the health of camera stream that are still recording data but somewhere in the video path portions of the video data is being lost.
You can use VSDI% as a reflection of various problems. If VSDI% approaches a value of 0, then that means that Viakoo has detected frame loss. Any value between 0 and 100 could be indicators that that you have or already have lost frames and may in danger of losing more for that particular stream.
A drop in VSDI can be caused by several events:
- Network latency issue
- Dropped packets or frames
- Storage latency (rising storage queue depth)
- Asset behaving above threshold
- Bandwidth usage
- File system corruption