AWS has announced Amazon File Cache, a new high-speed cache service on AWS designed for processing file data stored in varying locations. 

It gives applications access to using the POSIX interface, regardless of whether it is on-premises or on any file system that can be accessed through NFS v3 or on Amazon Simple Storage Service (Amazon S3).

File Cache creates a file system-based cache in front of either NFS v3 file systems or S3 buckets in one or more Regions and loads file content and metadata (such as the file name, size, and permissions) from the origin. It then presents this data to applications as a traditional file system. 

The solution helps with use cases where one’s application requires file access with sub-second latencies such as the management and transformation of video files, AI/ML data sets, and more. 

Users can link up to eight NFS file systems or eight S3 buckets to a cache and they’ll be exposed as a unified set of files and directories. The connection between File Cache and your on-premises infrastructure uses one’s existing network connection, based on AWS Direct Connect and/or Site-to-Site VPN. Users can also access the cache from various AWS compute services such as VMs or containers. 

There are no upfront or fixed-price costs when using File Cache and users are billed for the provisioned cache storage capacity and metadata storage capacity, according to AWS in a blog post.