In mission critical applications, the biggest challenge is how to securely deploy artificial intelligence (AI) from the cloud to the edge. When you’ve trained the AI and machine learning models in the cloud, the question at operational level is then how to deploy these cloud-based workloads to the edge and ensure security of the local system.
This is what the latest announcement from Lynx Software Technologies (Lynx) hopes to address. The company announced that its MOSA.ic platform now supports deployment of Google Anthos bare metal at the mission critical edge. The result is the ability to deliver containerized software services from the cloud, such as Google Cloud Visual Inspection AI service, providing a validated solution for secure, video-based quality inspection in industrial and energy facilities.
Lynx said it ensures the three functions – image capture (camera), insight via inference engine (Google Anthos) and the action with a supervisory controller – are completely sandboxed with the option of secure one-way (data diode) connections between them.
Through the partnership, the new solution enables:
- Easy deployment of real-time image capturing on devices such as cameras on manufacturing plant floors.
- Inference models built by Google Cloud Visual Inspection AI, that generate the insights.
- A supervisory controller that connects to the manufacturing execution system (MES) and translates insights into action.
In addition, Lynx’s technology brings immutable isolation and non-bypassable security to give plant or facility managers confidence the solution meets the OT (operational technology) security requirements.
In an interview with embedded.com, Pavan Singh, vice president of product management at Lynx, said, “A lot of focus to date has been on creation of AI models but very little maturity of deployment of AI and ML at an operational level where the local system is immune.” By supporting Google Anthos bare metal, this now means an entire Kubernetes cluster can be run locally in as little as one hardware system at the edge, with Lynx enabled virtual air gapping providing isolation between the different parts of the system.
“As a provider of mission critical edge, we’re thrilled to announce support for Google Anthos bare metal and Google Cloud Visual Inspection AI. With the new solution we’re jointly bringing to market, any containerized service can now be deployed to the mission critical edge without compromising security or performance. This partnership also marks an important step in our growing industrial ecosystem.”
Singh emphasized the benefits of providing this secure partition and Lynx’ experience in providing isolation, saying, “The separation piece is important. We’ve been on the case of separation kernels for about 10 years. The benefit of what we are doing is in being able to see the hardware, and also being able to run code on the bare metal.”
Lynx said for industrial and energy companies that are already feeling the strain of supply chain disruptions, labor shortages and more, this video-based quality system plays a significant role in enhancing performance and quality of output, while mitigating security risks. Through efficient visual inspection defects can be reduced by up to 10x, defective parts prevented from being shipped out, and insights can be gleaned into the cause of any defects to optimize processes. Singh said the visual inspection solution can open up more opportunities, with use cases such as part defect detection, welding seam inspection, PCB inspection, and silicon wafer defect analysis.
The solution is run on an Advantech MIC770, however, the Lynx MOSA.ic solution is able to run on various Intel and Arm processors. This flexibility means that various types of solution can have local/on-premise management while fully utilizing the benefits of cloud solutions.
- Why the future of embedded software lies in containers
- How containers improve the way we develop software
- Platform helps handle distributed edge apps as a single Kubernetes cluster
- Mobile edge computing powering virtualized 5G networks and industry