A forthcoming data caching service from AWS pits the market's largest cloud service provider against numerous hybrid storage vendors as it seeks to further lock users exclusively into its cloud ecosystem.
Amazon File Cache, a new service showcased at AWS Storage Day 2022 on Aug. 10, aims to bring hybrid cloud workloads deeper into the AWS ecosystem. The service creates a temporary, high-performance storage scratchpad in an AWS cloud region set by the user, and enables faster access and write speeds to NFS and S3 object data.
The service mimics the promise and technology of similar object storage systems using a file system front end such as Nasuni, Wasabi or Ctera, according to Dave Raffo, analyst at Evaluator Group, but it lacks some of the features that simplify setup and use.
"The difference is, those [vendors] have their own file systems, whereas with this you already need to use another NAS file system or on-premises file system," Raffo said. "It's adding caching to a file system, whereas the others are an all-in-one complete file storage program."
Cache and release
The Amazon File Cache service is still in development, but expected to be generally available before the end of the year, according to an Amazon spokesperson.
The service will be priced according to capacity and performance demands, like other AWS services, and can use data located in both on-premises file servers or object storage in the AWS cloud. Target industries for the service include media and entertainment or organizations that make use of computer-assisted drawings and renderings, both of which generate massive files and often involve multiuser collaboration.
Unlike other object storage services, AWS has no plans to support more advanced features such as file locking for collaboration. Data moving through the service isn't expected to be stored for extended periods, and those advanced features can be implemented using AWS Lambda, executives said.
At launch, the AWS service will be available in all regions, and supports NFSv3 file storage or S3 object storage -- both de facto standards for on-premises and cloud storage services.
AWS will continue to sell other file services with similar feature sets, such as FSx for NetApp OnTap.
All in one bucket
Although the service operates like other storage vendors' services to improve storage performance and access speeds, the end goal of the Amazon File Cache doesn't look to simplify hybrid cloud storage setups, said Ray Lucchesi, president of Silverton Consulting.
Every one of these storage vendors is playing the hybrid game. Amazon is trying to do the reverse of that.
Ray LucchesiPresident, Silverton Consulting
"Every one of these storage vendors is playing the hybrid game," Lucchesi said. "Amazon is trying to do the reverse of that."
Pushing more data through to AWS ultimately results in more user data residing in the AWS cloud and racking up charges, whether intentional or not, he said.
Running a caching service without an associated storage filer -- a common bundle in other vendor storage services -- could result in mixed performance for the service. "The problem with these caches is they have to index a lot of files that are out there," Lucchesi said.
The service is new ground for AWS to tread -- neither Raffo nor Lucchesi could recall a similar service sold by either Microsoft Azure or Google Cloud Platform (GCP).
Azure sells Azure File Sync, a service offering some of the unified namespace and cloud infrastructure capabilities of Amazon File Cache, but it can have difficulty keeping up with more demanding transfers and result in a slower user experience. GCP doesn't sell a comparable service.
Tim McCarthy is a journalist living on the North Shore of Massachusetts. He covers cloud and data storage news.
