Data is at the heart of most businesses, and is being generated at a dizzying rate. It has therefore become absolutely necessary that companies adopt effective storage management strategies. Having an efficient data storage management strategy is not about having all of your data highly available at all times. In fact, this approach will only result in your storage bills going through the roof.
Effective storage management entails utilizing the cloud’s different data tiers, so that your storage type, i.e. where the data is stored, matches operational requirements. Data tiering in Cloud Volumes ONTAP allows companies to do just this, by tiering data flexibly across storage tiers as needed. This functionality is a key element in efficient storage management.
In this blog post we explain what data tiering is, why it is needed, and explore how companies using Cloud Volumes ONTAP tier data successfully.
What Is Data Tiering?
Cloud providers offer many different storage types, each with its own latency, throughput, pricing structure, etc. As the value and usage of data can change over time, it may require leveraging the capabilities offered by one storage type over another. Data tiering is a method for moving data to the most optimal storage type at any given time.
Data can be tiered between storage types and storage classes. One option for a data tiering architecture is to use block storage as a performance tier for active, “hot” data, whereas object storage is relied on as a lower-cost capacity tier that houses inactive, “cold” data. Backup and disaster recovery copies for instance can be tiered to the lower-cost capacity tier and restored automatically when needed.
Another use case for data tiering is related to data lifecycle and moving dynamically between block and object storage. In the Healthcare sector, for example, a patient’s medical data is accessed frequently while they’re being treated and as such it needs to be stored in the block storage performance tier. However, once the patient’s treatment is complete, their medical data won’t need to be accessed much, if at all. But this data must be saved for the patient or for regulations such as HIPAA that require it be stored for a specific period of time. A solution to store this data efficiently is to move it to a lower-cost, block-based capacity tier until it’s needed again.
Cloud Data Tiering Challenges
Tiering data in the cloud between a performance tier and a capacity tier can be an effective solution for optimizing storage usage and costs, but one that comes with its own challenges. These include:
- Manual tiering is not efficient: Due to the massive amounts of data that are required to be stored, it is not possible to manually determine which data is hot or cold and then tier data between different storage tiers effectively.
- Lack of built-in support: While there are options for tiering between the storage classes offered by Amazon S3 and Azure Blob, there is no built-in support for automatically tiering between the block storage and object storage types.
- Footprint/cost concerns: The object storage capacity tier may be less expensive, but the cloud providers are in the business of charging for storage on a per GB basis—their storage services don’t offer built-in methods to store your data efficiently such as data deduplication, compression, or thin provisioning.
To solve these data tiering challenges, Cloud Volumes ONTAP has a powerful and automatic data tiering capability that allows data to move to the most appropriate storage type.
Storage Tiering with Cloud Volumes ONTAP: Going from Block Storage to Object Storage and Back
While cloud providers only support data tiering within storage classes of their object services, Cloud Volumes ONTAP enables automatic and seamless transfer of data from block storage to object storage both in AWS and Azure. Because the cost of object storage is considerably lower than that of block storage, using this functionality for cold data storage could dramatically lower data storage costs.
Cloud Volumes ONTAP offers three data tiering policies: Auto, Snapshot Only, and DR. These policies are highly flexible and can be configured for each volume:
Auto: With this option, cold data will be tiered to object storage, with a default cooling period of 31 days, or another period defined by the user. When data is needed again, Cloud Volumes ONTAP will transfer it back to block storage automatically.
Snapshot Only – Snapshot data is automatically tiered to object storage, with a default cooling period of two days. If the snapshot data is read, it will automatically be transferred back to the performance tier storage.
DR: This tiering policy is designed for disaster recovery and archive copies. With this option, data will gradually be moved back to block storage if activated in response to a failover or restore operations.
Data Tiering with Cloud Volumes ONTAP: Customer Case Studies
Companies from a wide variety of vectors have achieved major cost savings by tiering data with Cloud Volumes ONTAP. In this section we dive into three examples of large enterprises that use Cloud Volumes ONTAP’s data tiering functionality to optimize their storage usage and significantly reduce their cloud costs.
Reach PLC (formerly Trinity Mirror)
Reach is the largest news publisher in the UK, boasting millions of readers each month. It publishes over 200 print and online newspapers, including prominent brands such as the Sunday Mirror, Daily Mirror, Daily Express, and Sunday Express.
The company’s digital transformation began several years ago, when it adopted a hybrid cloud strategy, and began gradually migrating data to AWS. To facilitate this move, Reach partnered with NetApp. NetApp provides Reach with a data management platform from which it can manage content and drive efficiencies across all of its systems.
Cloud Volumes ONTAP, with its data tiering capabilities, helps Reach tackle the most important aspect of its business—data. Reach handles a massive and ever-expanding pool of data, including pictures, text and video, that must be processed in a fast-paced publishing environment. The picture desk alone receives over 100,000 daily images. With Cloud Volumes ONTAP, Reach is able to tier data to Amazon S3 storage, archiving data at half the cost of Amazon’s least-expensive disk option. This proved to be invaluable to the company, especially when it comes to its image libraries.
In addition to data tiering, Reach has managed to reduce its AWS storage costs by 50% by utilizing NetApp’s array of storage efficiency features, including compression, thin provisioning, and deduplication. The capital saved has exceeded the publishing giant’s expectations, and is currently used to invest in the company’s growth and development.
A leading multinational supplier of Ethernet and InfiBand Interconnect solutions, Mellanox Technologies provides its customers with solutions such as network adapters, processors, switches and cables designed to maximize business performance. The company runs multiple manufacturing sites worldwide and caters to a variety of sectors, including telecom, financial services, and network security.
On the path to digital transformation and reducing time to market, Mellanox was looking for a solution that could centralize its disparate data centers on one shareable and scalable platform. They chose NetApp for this task, because it demonstrated unmatched capabilities, and provided a much-needed hybrid solution, allowing Mellanox to seamlessly integrate their on-premises data centers with the cloud.
With Cloud Volumes ONTAP, Mellanox has managed to tier a whopping 90% of its cold DR and backup data to Azure Blob storage, Azure’s object storage service. While this reduces costs, it does not impede performance—with frequently-used data stored on SSDs and therefore easily and quickly accessible.
With Cloud Volumes ONTAP, Mellanox’s operations are now centralized on one scalable and cost-effective platform. Insights can be taken from its various factory sites, and changes can be introduced with agility. By using Cloud Volumes ONTAP’s built-in storage efficiencies, such as deduplication, Mellanox managed to reduce its production data footprint by 69%.
A leading Australian office supplies retailer, Officeworks has over 7000 employees and operates over 160 retail stores. After having moved some of its workloads to AWS, the company decided to refresh its legacy systems and embrace a hybrid cloud strategy that would optimize their business operations. With the help of NetApp Cloud Volumes ONTAP, Officeworks has begun to turn its cloud-first vision into a reality, breaking away from the constraints of its legacy data center systems.
Cloud Volumes ONTAP’s data tiering feature allows Officeworks to make the most efficient use of the AWS resources at its disposal. With Cloud Volumes ONTAP, the company can now automatically offload its infrequently used data to the lower-cost Amazon S3 storage, seamlessly transferring data back to performant AWS EBS as needed. Consequently, the company has not only lowered its AWS costs, but has also improved its operations—storing active data on high-performance disks while utilizing a lower-cost capacity tier for the vast majority of their data, which is in fact infrequently used.
With Cloud Volumes ONTAP, Officeworks was able to easily lift and shift their on-premises workload into AWS. Rather than continue to purchase physical storage for cold data, which must be regularly maintained, Officeworks has set up another Cloud Volumes ONTAP instance for its DR and dev/test functions. Since most of this data is cold, it is tiered to lower-cost Amazon S3 storage. Once data is in the cloud, it can be cloned with NetApp’s data cloning technology and sent anywhere. The data can also be made accessible through NFS, SMB/ CIFS, and iSCSI protocols with Cloud Volumes ONTAP file sharing features. Finally, Officeworks uses Cloud Volumes ONTAP with NetApp Trident to successfully create their persistent storage for Kubernetes.
Information is an incredibly valuable asset, and is being generated at a rapid rate. Companies that wish to thrive in such an environment must manage their data storage as efficiently as possible, making full use of the cloud’s different performance and capacity tiers. The best way to do so is by using Cloud Volumes ONTAP’s cloud tiering functionality, which allows you to seamlessly and automatically move data between tiers, using your cloud resources flexibly and as cost-efficiently as possible.