Edge Computing Partnerships: Edge computing is becoming ubiquitous. Companies have started understanding the benefits of moving towards decentralized computing. Edge computing brings in the concept of data analysis at the source of data generation that is at the edge of the network instead of the centralized location. It helps to reduce the enormous volumes of data generated by the IoT sensors from moving upstream and downstream of the system and thus increasing the latency and reducing the security of the data transmitted. As we know, lots of companies have come out with service, hardware, application for IoT edge. Research estimates put edge computing spending will reach $11.2 billion by 2024. Several of the key players have started to make significant announcements which include a new product, application, OS, partnership in edge computing to make their mark and claim their portion of the field in the edge computing battlefield.
Qualcomm has been researching AI for Edge. Their focus is on device AI at the wireless edge. Qualcomm has been focused on artificial intelligence for a long time. Now with the push and growth on edge computing and Qualcomm is trying to address the challenges of edge computing. The major bottleneck of bringing AI to edge is the compute resources. On a centralized data storage, one can implement AI, deep learning, and ML using all the resources available from the data center available for this. However, we are trying to bring this down to edge devices; the challenges become multifold.
With limited computer power, it is expected to provide a consistent user experience. The other challenge is the power consumption. ML, deep learning applications are power hungry; the edge devices are power constrained. Meeting the balance will be the answer to bring the AI to edge. Moreover, this where Qualcomm is bringing in Innovations by designing compact neural nets and optimizing the model for efficiency
Nvidia launched EGX an AI-enabled edge computing platform. They are calling this Realtime-AI at the edge. NVIDIA EGX is a scalable platform, allowing users to configure starting small with NVIDIA Jetson® Nano™ and scaling to full rack of NVIDIA T4 servers. NVIDIA EGX is compatible with NVIDIA AI computing offered by significant clouds. This means the application developed for the cloud can run on the EGX. They also have EDGE stack that can run IOT cloud services.
NXP and Microsoft have announced a collaboration aiming to deliver a New Microsoft Azure Sphere certified crossover application processor. This collaboration is to provide a secure and smart processor for IoT edge application that can run Azure Sphere Security platform. It will be based on Arm Cortex-A35 arch and will have powerful GPU and will consume low power running Microsoft Azure Sphere OS. The processor also has inbuilt security capable of generating and storing cryptographic keys
HPE in its Discover event this July has shown it’s moving ahead with an edge by announcing its strategy with an intelligent edge with enhancements to Aruba central a cloud-based platform. It will also likely make further edge related telco announcement. The joint statement by AT&T and HPE that the two companies plan to go to market with programs to accelerate edge computing. This partnership will combine AT&T Multi-access edge computing service with HPEs edge line converged edge systems to create the perfect environment for computation and processing at the edge
Akamai announced its plan to deliver service for IoT edge. They are launching IOT edge connect which provides access to MQTT (message queuing Telemetry Transport) and is capable of supporting millions of endpoints and can process ten times more message that the existing solutions. This IoT edge product will optimize the data delivery service and also reduce the power consumed by these devices.
We covered here some of the recent announcements on Edge Computing partnerships but not a comprehensive list. There has also been a host of other Edge Computing Partnerships announced by various other players this year. Industry experts have already defined 2019 as the year of the IoT Edge where Edge takes center stage of all IoT solution implementations. We are bound to see the IoT Edge stories grow.
- Mobodexter, Inc based in Redmond, WA builds IOT Edge solutions for enterprises applications on Kubernetes & Dockers.
- Register now at developers.paasmer.co and enjoy free Edge trial license that can run on Raspberry Pi.
- Follow our all our weekly IoT Blogs: http://126.96.36.199:8004
- Join our IoT Hub of 3000+ subscribers to get these blogs in Email – Subscribe.
- Become our affiliate partner today.