I am extremely excited to introduce the newest version of SwiftStack – SwiftStack 6. In this blog post I will talk about one of the headlining features of this release, namely the introduction of SwiftStack File Access.
Most know SwiftStack as private-cloud object storage for applications using the AWS S3 or OpenStack Swift API
SwiftStack has enabled our customers to deploy and operate private storage just like the public cloud, with support for the leading object storage APIs.
“Joe, we love SwiftStack, but I have applications that don’t use the object API”
While the growth in applications that targeted the AWS S3 or OpenStack Swift API has been awesome, there are a awful lot of applications that are out there that don’t use object APIs. We are extremely empathetic to teams who are trying to figure out how to modernize their applications – using public cloud or containers. These teams would _love_ to use object APIs, but have an important legacy of applications that utilize more traditional network filesystems protocols such as NFS or SMB.
SwiftStack 6 introduces integrated File Access
Go figure, when you have a robust method of storing and distributing data, it’s a great platform to build upon. SwiftStack 6 introduces integrated File Access that provides network filesystem access and object API access to the same data.
This architecture enables cloud-native applications using object APIs to have access to the same data as applications utilizing a filesystem.
Tastes great and less filling / have your cake and eat it too / best of both worlds – etc., etc.
Now applications and workflows that need the scale of a private cloud solution can leverage SwiftStack through network filesystem access without having to deploy a gateway solution.
For example, a media production can have a remote site capturing video footage. This remote production site can use an object-based client (such as the SwiftStack Client) to upload the data into SwiftStack over the internet (or private network). This avoids the need for specialized data transport tools. Then, once the data arrives in the cluster, an application that uses a filesystem – such as a transcoding application – can directly pick up the file and begin processing. Later on, finished media could be accessible via object APIs, and distributed via a content-delivery network that expect an object-based API.
During this entire workflow the same data is accessible via network filesystem protocols (NFS / SMB) and is accessible via object APIs (S3 / Swift). This saves an enormous about of time copying data between systems. And guess what, when any of those applications do end up supporting an object API — they are ready to go.
How did we do it?
We did a few very clever things, not the least of which was getting the right team working on this challenging problem. Look for an upcoming post discussing the architecture and how we are opening this to the community. With SwiftStack, anything that touches that data remains open source and utilizes open standards for accessing data.
How do I try it out?
To give it a shot go take a test-drive – https://swiftstack.com/test-drive/ Hit us up and we can enable File Access on your account.