Edge computing and edge artificial intelligence (AI) and machine learning (ML) applications are increasingly making a difference for organizations across multiple industries. Majority of consumer interactions still happen in stores, offices and theme parks which are edge locations. Similarly, many operational interactions take place in factories and warehouses which are also edge locations. At these edge locations, an increasingly vast amount of data is being produced through cameras, sensors and other IoT devices. This has made edge locations a fertile ground for deploying new AI/ML applications across retail, manufacturing, healthcare and even public sector industries.
In the healthcare industry, for example, CT scanners, video cameras and many other sensors are producing a significant amount of data. Because of consumer privacy and strict regulatory requirements, hospitals and healthcare organizations are keenly exploring use of Edge AI/ML applications to offer in-premise discoveries and findings. In factories, high-speed cameras offer a new way to detect and remove defective products before they reach consumers and lead to expensive recalls. As businesses increasingly rely on AI and ML to drive innovation and deliver enhanced user experiences, the need for robust edge computing solutions has become more apparent.
Enabling AI and ML at the Edge
Manjul Sahay, a product manager at Google, leads the development of the ‘Google Distributed Cloud Edge Appliance,’ an edge computing solution designed to help businesses deploy AI and ML applications in remote locations, even in environments with intermittent network connectivity.
The Edge Appliance offers an easy-to-use solution that integrates with cloud services, enabling enterprises to explore new possibilities and deliver unconventional user experiences, regardless of location. Google Distributed Cloud Edge Appliance is broadly in the same category as Azure Stack HCI and AWS Outposts. It is targeting enterprises who need a modern, cloud native platform to run data-driven, compute-intensive AI/ML workloads at the edge. Each Edge Appliance comes with a 16 core CPU, 64GB RAM, an NVIDIA T4 GPU, and 3.6TB usable storage. The appliance also has a pair of 10 Gigabit and 1 Gigabit Ethernet ports. With the 1U rack-mount form factor, it supports both horizontal and vertical orientation which makes it well suited for various edge deployment scenarios.
Sahay led a team to develop Google Edge Appliance by focusing on key advancements in the area of security, AI/ML platform and network connectivity. First, Sahay and team focused on security features because security and data privacy are the most important concerns for any edge technology. Edge appliances are commonly placed in non-secure locations as against a corporate datacenter with multiple layers of security. An appliance placed in a retail store or a hospital or a factory can be easily removed by a determined miscreant. Appliances are also at the risk of tampering when being shipped to and from these locations. Sahay’s team identified and built a set of advanced security features such as dual security key encryption, storing the key in Trusted Platform Module instead of disk and shipping with a hardened Operating System.
Next, for the AI/ML platform, Sahay led the team to evaluate options and decided on using Kubernetes as the underlying platform. Sahay explained that using Kubernetes reduces the platform footprint, and makes available most of the appliance’s computing power for customer applications. Based on CPU benchmark and sysbench tests, Kubernetes resource utilization of CPU and RAM is up to 50% lower compared to the traditional VM approach. Also Kubernetes applications can be easily built on cloud and then quickly deployed across tens or even hundreds of edge appliances. This Kubernetes platform choice is different from that made by other large cloud players like Amazon and was a bold decision for the team.
The final challenging piece was solving for intermittent network connectivity. Sahay built on previous work of Storage Appliances to incorporate the technology which makes these appliances work without network connectivity for up to 90 days. Despite intermittent network connectivity, the appliance’s ability to operate with full AI and ML capabilities differentiates it from traditional edge computing solutions. This resilience is crucial for businesses operating in remote or challenging environments where reliable connectivity is not always guaranteed.
Multiple Applications Across Industries
Edge Appliance has potential AI/ML applications across various industries, including telecommunications, manufacturing, autonomous vehicles, and retail. Sahay’s Google Cloud team has focused on a few leading use cases and customers for these Appliances. One such use case is Advanced Driving Assistance System (ADAS). Nuro, a leading innovator in the ADAS space, collects a lot of data from remote environments like depots which may not have continuous connectivity. Once the data is collected, it can be processed locally or transferred to the cloud using these appliances.
Another use case is for retail sector customers to overhaul store management operations, for example, monitoring store occupancy, queue depth and wait times, detecting slips and falls and monitoring inventory compliance. Sahay and team are working closely with these customers to build industry-specific customizations such as supporting automotive video transfer protocols for appliances sitting in a car or rover and enabling defect detection applications on the appliances for manufacturing customers.
Security Concerns, Industry Sentiment and Impact
While the Google Edge Appliance has generated excitement within the industry, some critics have raised concerns about the potential security risks associated with deploying AI and ML applications at the edge. Cybersecurity experts are worried about privacy implications of extensive data collection at edge, data theft, ransomware attacks, DDoS (Distributed Denial of Service) attacks which can now target hundreds or even thousands of these devices placed in an organization. In this context, Sahay’s work on dual key encryption and hardened OS security features has significantly improved the security posture of these appliances.
Despite concerns ,overall sentiment within the industry remains positive, with many experts predicting that edge computing will play an increasingly critical role in the future of business -technology. Sahay’s work in building and customizing Edge Appliance has potential to significantly impact multiple industries. Supporting growth and development of ADAS is itself a game-changer. Similarly, retail and manufacturing companies significantly reduce expensive store operations problems and manufacturing defect problems, potentially reducing costs by 10-20%. Telcos can create new revenue streams. Appliance unlocking new opportunities for innovation and growth across various industries making it easier for businesses to deploy AI and ML applications at the edge. Early customer success indicates an impactful future.
As businesses face the challenges and opportunities presented by technology’s rapid changes, product managers like Sahay help shape the future by helping businesses adapt and succeed in an increasingly complex world.
Read More From Techbullion
