Innovation in Surveillance: What’s Changing at the Edge, Core and Cloud?
What’s new in the surveillance market? To say that interesting things are happening around surveillance technology would be a huge understatement. As we gear up for ISC West this year, some of the most significant trends we’re seeing are: artificial intelligence (AI), machine learning, the power of IoT, combining different data sources to add meta data and layers of context coming to the surveillance market and pushing for faster evolution of the edge of surveillance systems. All of these trends have been exploding over the past 18 months. As these trends drive change and innovation, what’s changing at the edge, in the core and on the cloud?
More Storage at the Edge
It is mission critical for a system to be able to collect video whether the cameras are connected to an NVR, to edge storage units, or on the individual surveillance camera itself. Collecting the data is the basis of a surveillance system and is the most critical point of failure in many implementations. Relying on the always-on connectivity between the camera and a recording device is a risky strategy. What happens if the network is not available due to a technical issue, weather conditions, or a deliberate attack?
A number of failsafe mechanisms can be deployed – from using multiple cameras to cover overlapping areas, RAID solutions on the main storage, or memory cards at the individual cameras and distributed storage systems. Storing data inside a surveillance camera is often used as a failsafe mechanism in the event that connectivity to the NVR or cloud is lost.
Real-Time Analytics at the Edge
Today, with the commoditization of compute power and advancements in flash technology, cameras can do more than just capture streams to feed to an NVR or, in the case of an enterprise, transfer all the footage to the cloud at the data center. Today, more compute power is being driven to the edge. We’re also seeing the evolution of more data analytics happening at the edge. Applications such as facial recognition, number plate identification, and object classification can happen at the edge providing real-time data and insights on the situations as they are being captured.
What Happens at the Core and on the Cloud?
The fact that more horse power is available at the edge does not take away from the importance of other components in the system. Today’s data environments are more distributed. Video is no longer stored in a single node. Even though the data environments remain focused on capture, store and analyze, data is not flowing in a single direction.
As mentioned above, it is becoming common to see data analyzed at edge, near the source or capture device. This brings time-to-insight advantage, saves on the cost of communication and can be used as a failsafe precaution. However, many system will still deploy a centralized NVR to act as the core of the system for reasons of privacy/security and for cost-effectiveness. Even if the system will archive data to the cloud, the NVR might still be used as a local gateway, making sure the data that moves to the cloud is not sensitive data.
In addition, there is still a need to collect large amounts of data to be able to observe patterns that can be used for model training, or training the AI engines. This can be done either in the core of the system or on the cloud. There is also a need in many systems to provide long-term archives of footage which can also be done in the core, on the cloud or distributed between the two.
The structure of data storage is now a hybrid solution of fast data stored and accessed close to where the action is and slower big data residing closer to the core of the network. Finally slow/cold data is eventually stored on a storage system optimized for archival or on the cloud for long term retention and usage.
Beyond Surveillance… Getting More Value from the Data at the Core
In surveillance systems huge amounts of data are collected, but often very little of the data is utilized. The data collected is analogous to an insurance policy. Unless a security incident occurs, you do not need the data, or the insurance. However, the data collected can be extremely valuable for other types of business insights that are not about surveillance and security but about commerce and other location/context-based applications.
Smart video takes surveillance data beyond simply collecting data to providing actionable insights for applications such as agriculture, building construction, Smart Cities and oil & gas. We’ll continue to explore these applications in future blogs.
Come visit us at ISC West next week in Las Vegas to learn more about Western Digital’s solutions for the surveillance industry.
Certain blog and other posts on this website may contain forward-looking statements, including statements relating to expectations for our product portfolio, the market for our products, product development efforts, and the capacities, capabilities and applications of our products. These forward-looking statements are subject to risks and uncertainties that could cause actual results to differ materially from those expressed in the forward-looking statements, including development challenges or delays, supply chain and logistics issues, changes in markets, demand, global economic conditions and other risks and uncertainties listed in Western Digital Corporation’s most recent quarterly and annual reports filed with the Securities and Exchange Commission, to which your attention is directed. Readers are cautioned not to place undue reliance on these forward-looking statements and we undertake no obligation to update these forward-looking statements to reflect subsequent events or circumstances.