Cloud, AI and Kubernetes: The Transformation of Physical Security
I’ve seen tremendous product advances in my 17 years in the video surveillance industry. Along the way, I’ve also heard lots of claims that fall short. While many companies are well intentioned, technologies often don’t perform as advertised. Or multiple dependencies, high costs, and massive compute and bandwidth requirements in real-world settings make them impractical. It’s challenging for integrators, resellers, and end users alike to have the technical knowledge to recognize the shortcomings before it’s too late. And the entire industry suffers as a result. But every so often there is a foundational shift — like the move to cloud — that changes everything.
Spending the past four years building a VSaaS platform from the ground up. I realized quickly that our industry is entering a new era — similar to moving from analog to IP and from low-resolution cameras to high-resolution megapixel units. Cloud coupled with Artificial Intelligence (AI) and Machine Learning (ML) have turned data into actionable intelligence that is reshaping physical security. Now billions of dollars invested by the tech giants to manage their own public cloud services has transformed the TCO of on-prem systems too. We’re on the brink of another platform shift that turbocharges the benefits of cloud and AI, and makes them vastly more economical, scalable, and effective.
Kubernetes: the next big platform shift
Modern applications are increasingly built using containers, which houses the code, dependencies, and environment in one logical block. Enterprises are using containers to increase efficiency and resource utilization, improve security, automate workflows, and accelerate innovation. It’s not surprising that Gartner predicts more than 75% of global organizations will be running containerized applications by 2023. Kubernetes is the leading open-source platform for managing containerized workloads and services. Using Kubernetes, AI can run anywhere — across different platforms, toolsets, and chipsets — and scale instantly.
A compounding effect on product quality
In addition to the ability to deploy anywhere and execute AI models at any place in the video processing pipeline, the advantages brought by fast innovation have a direct result on product quality. With Kubernetes, there is no more waiting months for slow release cycles to deploy new capabilities. The open interoperability removes gating dependencies and technology blockers so organizations can deploy new capabilities and scale exponentially faster. Open interop brings a continually expanding ecosystem of continuous improvement. We’ve seen examples of the wholesale product category replacement resulting from platform shifts that radically change the rate of innovation. In the early 2000’s, the Motorola Razr phone enjoyed huge popularity for its sleek styling. Yet, it was quickly overtaken by iPhone and Android models because of their ecosystem power.
Up to 90% lower TCO
AI is often shown as the “bells and whistles” of video analytics, but its biggest value is in the “plumbing” that reduces total cost of ownership (TCO). For example, choosing the video encoding and storage location dynamically based on the activity observed (human in restricted area vs. passing car in the distance) often saves 80% on storage alone. And deploying AI through Kubernetes reduces TCO even further, by up to 90% in some cases. Cloud-native containerization requires less hardware and therefore less maintenance. IT control of storage, authentication and lifecycle management reduces costs and delivers efficiencies. Instead of manually configured OS, storage, software updates, and security patches, upgrades are fast, automatic and don’t interrupt services.
Natural language video search for real-time alerts
In this new environment, companies with cloud and AI baked in rather than bolted on have a distinct advantage. Camio is an example. Camio’s cloud-native architecture driven by Kubernetes applies AI throughout the video processing pipeline for actionable business intelligence at a very low cost. Natural language video search becomes a real-time alerting platform. Any pinned search becomes an alert instantly. Interactive, customizable dashboards show up-to-the-minute data as well as long-term trends that inform improved security and business operations. I can’t help but be excited by the power of a platform that’s so easy to use and brings so much more value to customers than traditional security solutions built as server applications for on-premise use only.
A future in the cloud
Over the next several years, cloud-native systems will rapidly replace standalone server applications. AI applied via cloud-native systems doesn’t just speed forensic investigations, it deters and detects threats proactively and increases ROI. From a business perspective, the ability to use the same system that protects people and property for other operational insights like detecting safety equipment or capturing customer behavior for better retail marketing decisions opens up new opportunities through shared budgets. The companies that use cloud-native Kubernetes not only draft off of the big tech investments in lowering TCO but also innovate faster.
I feel fortunate to help people on their path amid this tremendous change. As a life-long lover of technology, I’m energized by the opportunity to educate customers and guide them on their choices. No one solution or provider will be right for every company and every opportunity. But I believe customers will become a much more educated buying force. They will demand more from their video surveillance systems and be better equipped to make smart AI choices and transition to the cloud.
— Hunter Fort, Vice President of Sales at Camio.
