Today, we announced version 5 of SwiftStack Storage to help our customers further realize the value of cloud storage on-premises and to enable new hybrid cloud workflows. So let me run through the details…
You may ask “Why 5 and not 5.0?”. Unlike most traditional enterprise storage products, we frequently roll out SwiftStack software updates, allowing our customers to quickly take advantage of improvements and new capabilities. So what you’ll see very soon is a SwiftStack 5.0 release that will have some new capabilities in it and then 5.1 release a few weeks later with additional capabilities and so on. When we make product announcements like this, think of it as a preview and what’s to come from us in the near term. For example, the announced SwiftStack 4 features last year were delivered in the first few 4.x releases.
SwiftStack Cloud Sync
SwiftStack 5 will enable the ability to bidirectionally sync data with Google Cloud Storage and Amazon S3. So now, data can be replicated back down to on-premises storage from a bucket in the public cloud. Customers have the flexibility to choose one-way or two-way sync on a bucket by bucket basis depending on their need or use case.
Uses we’re focused on for Cloud Sync are:
Data Protection – If you do not have a second data center to utilize to protect your data from a major disaster, the public cloud is a viable alternative. Cloud Sync automatically replicates your vital business data offsite. If needed, two-way sync allows you to easily restore data on-premises.
Collaboration with Others – If you need to collaborate with outside teams, deliver assets to a client, or utilize an application running in a public cloud, but for whatever reason cannot provide access to your private cloud storage, data can be made available in a public bucket.
Cloud Bursting – You truly start to take advantage of hybrid cloud storage when you can utilize the elastic compute capabilities of the public cloud. Cloud Sync keeps data in a cloud-native format, allowing applications to act on the data.
Offsite Archiving – Once your data exists in the public cloud, you now have the option to leverage cost-effective cold storage, like Amazon Glacier or Google Cloud Coldline, for offsite archiving. Two-way sync allows that data to easily come back on-premises if need be.
While applications can read and write data into SwiftStack using object APIs, end users also need easy access petabyte scale storage. We’ve built an all-new SwiftStack Client, which runs on macOS, Linux, and Windows, which provides simplified access to SwiftStack containers/buckets.
Once you log into your SwiftStack accounts, containers/buckets you have access to are automatically shown. You can then browse, search, write data into, or read data out of those containers/buckets using the simple GUI interface.
Multi-Region Erasure Coding
SwiftStack can protect data by replicating it across failure domains or using erasure coding. Until now, using erasure coding was limited to a single data center. SwiftStack 5 now has the ability to use erasure coding across multiple data centers or multiple geographic regions.
The choice of replicas or erasure coding and levels of protection are part of a storage policies, which are applied to specific buckets/containers. This means that different data protection methods and geographic locations can be used for different data sets in the same SwiftStack cluster.
While it’s been a recommendation to limit the number of objects in a containers/buckets to a few million, SwiftStack supports a near-unlimited number of containers/buckets. New in SwiftStack 5 will be the ability to store hundreds of millions of objects in a single container. This improves compatibility with applications written to use Amazon S3 that assume infinitely deep buckets.
SwiftStack Controller Software Appliance
The SwiftStack Controller is the single pane-of-glass to manage your SwiftStack storage clusters. It’s a SaaS application in the cloud for ultimate simplicity or can run on your infrastructure behind your firewall. When you do decide to deploy it on-premises, it is now very easy to install and maintain on hardware of your choice or a virtual machine.