Cloud Volumes ONTAP extends the trusted enterprise data management capabilities of NetApp to leading cloud platforms Azure and AWS. In Azure, it provides CIFS/NFS/iSCSI-based services to hosted workloads, leveraging the underlying Azure storage and compute resources with additional data storage economy through advanced storage efficiency features such as thin provisioning, deduplication, compression, and now tiered storage to Azure Blob storage.
Cloud Volumes ONTAP’s new data tiering in Azure is another feature that enables efficient storage management, tiering out cold data to a cheaper storage tier. Storage tiering brings in storage economy with minimal operational and management overhead. This blog will explore the storage tiering feature available in Cloud Volumes ONTAP for Azure storage and how customers can greatly benefit from it for any type or use of their data: active data, snapshot data, DR data, or any other dataset.
What Is Storage Tiering?
Storage tiering is the ability to keep data in different pools of storage as needed. Since different storage volume types have different characteristics, such as latency, throughput, and access charges, they also are priced differently. A capacity tier is used to house data that is not needed for frequent use at a lower cost, and a performance tier for all the data in active use.
In Azure, the capacity tier would be on Azure Blob and the performance tier on Azure disks. Automated storage tiering in Cloud Volumes ONTAP provides a huge benefit by allowing users to automatically and seamlessly move data between the capacity and performance tiers as needed, depending on the data frequency usage, something the native cloud providers can’t do.
Overview of Cloud Volumes ONTAP Data Tiering in Azure Storage
While storing data in the Azure cloud, storage is charged based on usage. Hence it is sensible to move infrequently-accessed data—such as snapshot data, DR environments, and backup copies of data—to a cheaper storage tier to improve cloud storage costs. Cloud Volumes ONTAP eases this task with the new data tiering feature, where infrequently-accessed data, i.e. the “cold” data, is automatically and seamlessly tiered to an Azure blob which acts as the capacity tier, and this can be done with minimal configuration overhead. Data frequently accessed by applications, or the “hot” data, remains in the performance tier, which is located either on the Premium SSD managed disks or Standard HDD managed disks service in Azure.
Azure Blob Tiers
Now let’s look at the tiers that Azure offers for Azure Blob. Azure Blob storage offers three storage tiers currently: hot, cool, and archive. (Note that these are different than the hot and cold tiers of Cloud Volumes ONTAP parlance; Cloud Volumes ONTAP considers all of Azure Blob a cold tier). Of these, Cloud Volumes ONTAP uses the Azure Blob hot and cool tiers. The Cloud Volumes capacity tier defaults to Azure Blob hot tier, and data is moved to the Azure Blob cool tier if it is inactive for 30 days. Cool tier is cheaper compared to hot tier, but is intended for data that will remain in the tier without being accessed for the minimum 30 days. Though cool tier costs less than hot tier, data access charges are higher and hence customers should analyze the data being stored and access patterns before changing tiering levels.
The tiering is done automatically based on the data usage and hence there are no additional monitoring or configuration overheads. By leveraging cost-effective storage, the monthly storage consumption bill can be reduced considerably. For example, according to Azure storage pricing, storing 1 TB of data in Azure cool tier with minimal read/write will cost approximately $18/month. However, the same data when stored in a Premium SSD storage in the performance tier can cost you approximately $77/month. The annual cost savings in this example is $216, which translates to approximately 75% cost savings with no overhead when compared to the cost of Premium tier storage.
Cloud Volumes ONTAP Data Tiering Policies
Four tiering policies can be associated with a volume when you create, modify, or replicate it. Tiering policies are independent of volumes and you can apply different policies per volume. The policies are as follows:
Snapshot only: Snapshots are online, point-in-time, read-only copies of the data in a storage volume. If the Snapshot only tiering policy is selected, the cold snapshots are tiered to the capacity tier after a cooling period of approximately 2 days. The performance tier is thus freed up.
Auto-tiering: If data is not accessed for 30 days or another period as specified by the user, it is automatically tiered to Azure blob storage. If the data is again accessed after moving to the capacity tier, it will be automatically and seamlessly tiered back to the performance tier, i.e. Premium SSD or Standard HDD managed storage. Cloud Volumes ONTAP has the intelligence to detect sequential data reads such as antivirus scans or file indexing, and in such cases, data is not tiered back to performance tier.
Disaster Recovery: With volumes replicated for the purposes of DR, the destination volume is created by default in the capacity tier. If this replicated volume is later accessed for say recovery operations, the data is moved to the performance tier as just as when it is accessed normally.
None: In this case, data remains in the performance tier and is not tiered automatically.
Note: Both snapshot only and auto-tiering policies kick in only when the capacity reached an aggregate of 50%.
Storage Tiering Prerequisites and Configuration
Storage tiering is supported with Cloud Volumes ONTAP starting with version 9.4. No additional licenses are required. The configuration steps for tiered storage architecture can be summarized as follows:
It is important to ensure connectivity between the capacity and performance tiers to enable Cloud Volumes ONTAP storage tiering. However, if Cloud Manager is already configured with the required permissions in Azure, this requirement will be taken care of automatically. Cloud Manager will also configure the required VNET service end-point necessary for tiering to work.
Cloud Volumes ONTAP is helping customers all over the world in their digital transformation journeys by facilitating Azure storage management and empowering storage economy. The success story of Galatz Radio is worth taking a look at, as they achieved as much as 70% savings in data center and storage costs by leveraging Cloud Volumes ONTAP for Azure. The new tiered storage solutions in Cloud Volumes ONTAP further enhances the savings potential by leveraging low cost storage for infrequently-accessed data. If you’re also working in the AWS cloud, find out about storage tiering in AWS here.
Organizations planning to use Cloud Volumes ONTAP for enterprise use cases such as backup, DR, as well as any other workloads would greatly benefit from automated storage tiering with Azure. If you’re part of one of them, try a Cloud Volumes ONTAP 30-day free trial on Azure today.