An extension of the AWS cloud within the telco environment, AWS Wavelength allows developers to build ultra-low-latency applications for mobile devices and users by deploying AWS compute and storage at the edge of Verizon's 5G network. The first two AWS Wavelength Zones (AWS infrastructure deployments that embed AWS compute and storage services within the telecommunication providers’ datacenters at the edge of the 5G network) are available in Boston and the San Francisco Bay Area with additional locations planned for later this year.
By embedding AWS compute and storage services at the edge of 5G networks, says the company, AWS Wavelength enables developers to serve edge computing use cases that require ultra-low latency like machine learning, Internet of Things (IoT), and video and game streaming.
"Our customers tell us they are excited to build applications that take advantage of 5G networks," says said Dave Brown, Vice President, EC2, Amazon Web Services, Inc., "but in order for these applications to provide ultra-low latency to mobile end-users, customers need AWS compute and storage services embedded directly within the 5G network. With AWS Wavelength, our customers can develop applications that take advantage of ultra-low latencies to address use cases like machine learning inference at the edge, smart cities and smart factories, and autonomous vehicles - all while using the same familiar AWS services, API, and tools to deploy them to 5G networks worldwide.”
With 77 Availability Zones across 24 AWS Regions, says the company, AWS currently enables developers to serve end users with low latencies worldwide. However, there are an increasingly a number of applications - e.g., smart manufacturing, machine learning inference, autonomous driving, live event streaming, IoT, and video games - that want very low latency across the mobile network.
To perform their jobs, these applications need access to compute and storage, and like many applications today, they also seek to leverage the cloud. However, accessing the cloud using traditional mobile architectures requires