In my blog post on the new features of clustered Data ONTAP 8.2, I highlighted three key benefits: nondisruptive operations, proven efficiency and seamless scalability. This blog explores seamless scalability in more detail.
A clustered storage environment running Data ONTAP scales in three dimensions: capacity, performance, and operations. You can scale SAN and NAS capacity from terabytes to tens of petabytes transparently and without reconfiguring running applications. Performance scales linearly as cluster nodes are added; a single administrator can manage petabytes of storage. You can start with a single cluster node and expand your cluster up to 24 nodes as your business needs grow.
Figure 1, Clustered Data ONTAP scales in several dimensions.
The nondisruptive-operations capabilities of clustered Data ONTAP make your storage environment much more flexible. You can easily rebalance capacity and storage workloads as needed. You can improve service levels by dynamically redeploying workloads and avoid hot spots by moving volumes to less active disks or spreading workloads across more controllers. Each dataset gets the right technology to meet your performance and cost targets.
In clustered Data ONTAP 8.2 we’ve also significantly increased many of the limits to make the platform even more scalable. This includes support for:
- Up to 100,000 NFS clients
- Aggregates up to 400TB in size
- 12,000 volumes in a 24-node NAS cluster
- 49,152 LUNs in an 8-node SAN cluster
- Multiple Infinite Volumes, each up to 20PB in size
- Microsoft BranchCache v2, which allows Windows servers to cache data from a NetApp cluster locally to improve performance
A More Flexible Approach to Scale-Out
Most scale-out storage solutions provide a single large repository with limited control. Essentially, this means a single class of service for every workload sharing the storage. The virtualized storage services and unique capabilities of clustered Data ONTAP give you much more flexibility and greater control. With NetApp Infinite Volume, you can have a single repository or several large repositories to match your requirements, each capable of scaling to billions of files and petabytes of capacity. Any SVM can include an Infinite Volume, and you can control and adjust the resources available to the SVM—and the Infinite Volume—on the fly.
Clustered Data ONTAP gives you the ability to isolate particular workloads or tenants and offer different levels of service. You can mix certain controller types, offer different storage tiers, and define QoS policies to address a wide variety of storage needs, all from the same unified infrastructure. You can also virtualize third-party arrays with NetApp V-Series, and when it’s time to retire a storage system, you can simply upgrade the controllers—keeping data in place.