A look at scale-out NAS – Infinite Volume in clustered Data ONTAP

These days when speaking with customers about scalable storage infrastructure, the conversation often boils down to the three aspects of scalability: capacity, performance and operational scale. You can read more about them in Mike McNamara’s blog: https://communities.netapp.com/community/netapp-blogs/sanbytes/blog/2013/07/19/seamless-scalability-with-clustered-data-ontap.

Today, I want to focus on one aspect of scale-out – capacity. Clustered Data ONTAP enables customers to scale to 69PB of capacity in a single cluster. With Infinite Volume – which we first introduced in 2012 with Data ONTAP 8.1.1 – you can now scale to 20PB in a single volume. So Infinite Volume gives you a big, 20PB bucket to store data, along with the operational scale and performance you need for large content repositories. The unique thing, however, is that you don’t lose key functionality that enterprise IT departments expect. Essentially, Infinite Volume works just like any other volume from an end-user or application perspective, and in many ways from an administration perspective as well.

Let’s take a closer look:

Administration – you create Infinite Volume just like any other volume. Typically you would use a GUI wizard, but you can use the command line, too.

Efficiency – deduplication and compression are supported. You can even steer data via policies into separate storage classes that compress or deduplicate data, so you get to choose how data should be handled. Or let it all go into one single repository, your choice.

Scalability – well, 20PB of capacity, 2B files – and all that with only 10 nodes. So you don’t need dozens (or 144) nodes to get to this capacity, which saves you a lot of configuration and equipment headaches. This gives you a highly capable scale-out NAS solution, with support for NFS, pNFS and SMB/CIFS for data access.

Availability – the 99.999% uptime you get from Data ONTAP, even while you expand or manage an Infinite Volume or perform software updates. High-performance snapshots and replication via SnapMirror are also supported to increase data availability.

Multi-workloads – you are not limited to having Infinite Volume for large-scale content repositories in a cluster. You can easily add other workloads such as virtualized servers and desktops, enterprise applications and many other items while the system is running. Infinite Volume supports NFS (including pNFS) and SMB/CIFS for data access, but you can add other protocols like FC, iSCSI, FCoE etc. into the cluster for other applications. Great for large organizations and – with secure multi-tenancy supported – service providers that want to share some of that storage infrastructure with various customers.

In summary, with Infinite Volume you have the ability to easily create a scale-out NAS content repository for up to 20PB of data, while retaining efficiency, availability and many other aspects of clustered Data ONTAP – without excluding other workloads.

So don’t create another silo – leverage clustered Data ONTAP for all your workloads, even those where in the past you might have chosen another, often inferior, scale-out NAS solution.