A bank, for example, knows if it had a major incident relatively quickly, so having a week’s worth of video immediately available is OK. But balanced against that, there are also federal and legal mandates to keep data for many months to even many years in some cases. For long-term storage, it is senseless and wasteful to spend all that power and energy to keep hard drives operating.
There are a few great companies on the market today that will allow you to have up to 75TB of storage (9.5 years of full HD video) ‘online’ for less than 60W of power. Should something go wrong, the IP video management software (VMS) – with just a few mouse clicks – automatically spin up the right hard drive, grab the video and let you view it.
This application isn’t only for banks … this type of storage could be used at far less exotic customers than banking and critical governments. Restaurants, for example, are routinely sued by people claiming to have slipped and fallen or were injured in incidents that go years back. New York has a statute of limitations of three years. You can be sure that most restaurants don’t have video logging that far back, yet get lawsuits right about the time the limitation timer expires.
Examples like this prove that oftentimes, end users don’t understand or realize there are ultra-long-term storage options out there now. As such, it’s a great excuse to reach out and educate them and be able to provide value-added solutions.
Integrators are also co-opting local datacenters, creating their own pools of mass storage they can then lease (RMR!) to their customers. This makes a lot of sense since an integrator can tailor the service directly for their customers in the form of a real value-added service.
Amazon also has a cold storage solution called Glacier. It’s low-cost (less than 0.01 cents/GB per month), but the catch is you have to wait several hours to be able to retrieve your video. For storage, it can be cheaper than buying your own hard drive arrays, but there are costs to push and pull video into and from their cloud.
Storage is becoming almost free. Powering and migrating data into and out of storage, however, is still somewhat expensive.
Advanced Analytics Are an Answer
Video is really boring. Many people have installed cameras in their homes, only to check them periodically at the beginning out of curiosity before boredom sets in and then they are never viewed again.
Video is extraordinarily bad at conveying useful information, forcing the user to sit through hours and hours of video looking for something meaningful. Video content analysis has long promised redemption, but except for very specific applications it has never failed to disappoint. That experience is slowly changing now for a number of reasons.
First, thanks to the consumer electronics industry, the average IP camera is remarkably powerful … far more so than the unitasking job to which it has been enslaved. It’s time for modern IP cameras to come out of the closet and be tasked with greatness as well a
s be accountable and acknowledged for the little super computers they are. As an aside, many if not all of the issues and abuses with IP cameras from a cyber-threat perspective are because users are so dismissive of what these devices are capable of. No right-minded professional would put an unpatched server directly on the open network, but this is routinely done with IP cameras.
Second, the expectation of what end users want out of analytics has become far more realistic. Gone (hopefully) are the days of vendors pushing “terrorist detectors” and/or “find the backpack full of explosives.” The value now with analytics comes mostly in marketing and figuring out why someone bought something, or more often why they didn’t buy something. Now, that’s more reasonable and doesn’t have the same amount of potential for catastrophic liability that the aforementioned targets do. No one will lose their life if someone doesn’t buy a sweater.
