hamburger icon close icon

On-Prem Plus AWS?:
Using Amazon S3 in a Hybrid Cloud Infrastructure

When virtualization first started stepping into the data center, IT teams got a chance to leverage many benefits such as applications that were independent of hardware, easily achieved high availability, and lower costs through hardware consolidation. That same logic applies to using the cloud.

Consider what the cloud is: basically, it’s a bunch of hardware shared between different customers. There is really another data center out there in the world and spreading your data between that data center in the cloud and yours on-prem, is called a hybrid cloud infrastructure. But how can you do that? 

NetApp’s new Cloud Tiering service is a solution which can help AFF and SSD-backed FAS users begin to take advantage of that shared hardware by automatically moving their inactive data from high-demand NetApp storage in the data center and storing it in inexpensive object-based cloud storage service. On AWS, that service would be Amazon Simple Storage Service, commonly known as Amazon S3.

What Is a Hybrid Cloud Infrastructure?

Hybrid cloud infrastructure is when a user combines an on-prem data center with storage and compute resources that are available in the cloud. The main objective in these types of architectures is to leverage the cloud in order to save on costs for compute and storage equipment and operations.  

Because many companies are still not ready to go all-in with the cloud, especially at big enterprises that have a lot of legacy applications, hybrid infrastructures are some of the most popular models in the cloud today. Think of the advantages: operations get the ability to use pay-as-you-go cloud resources while still keeping their data close by in the data center.

Obviously, one major barrier to working with this type of architecture is getting all the parts to play well together. How can data be moved between the separate deployments? Is there a difference in performance that’s acceptable? And most importantly, does it involve having to rearchitect existing applications? That could wind up taking extremely long amounts of time and resources. 

There are different ways that hybrid environments can be created, but not many solutions which can help make sure that the data shared between the on-prem systems and the cloud can be actively used at the same time. NetApp’s Data Fabric approach can help, using a number of solutions. Some of these are well-known and already widely used in the on-premises world, such as NetApp SnapMirror® replication technology, which is used to replicate data to the cloud for active use or for disaster recovery and backup reasons. Another method that isn’t as well known is the new Cloud Tiering service for NetApp AFF systems and Hybrid FAS systems with SSD drives.

Cloud Tiering to Amazon S3 as a Hybrid Solution

Cloud Tiering is a managed service for tiering infrequently-accessed data from NetApp’s All-Flash FAS or even Hybrid FAS with SSD drives to object storage in the cloud. Cloud Tiering uses NetApp’s FabricPool technology to transparently move that data to the cloud, helping customers free up storage space in their data centers without having to make any changes to their applications.

Data that Cloud Tiering moves is stored in cost-effective object storage format. On AWS, that means to Amazon S3. Amazon S3 is an industry-leading, scalable, highly durable, secure, and performance-capable object storage service. Many companies use Amazon S3 to store their backups, for disaster recovery (DR) purposes, and as a data archive. Amazon S3 is easily managed and simple to use and has fine-tuned access controls to meet business and compliance requirements.

Amazon S3 offers a high level of availability, by automatically creating and storing copies of all S3 objects across multiple systems. This gives the service up to 11 9’s of data availability (99.999999999%), which is pretty great.

While affordability and reliability are the main draws for using Amazon S3, AWS has loaded it with plenty of additional features too:  Amazon S3 Cross-Region Replication helps replicate data to into other regions, Amazon S3 Object Lock for retention policies, and Amazon S3 inventory to get visibility into stored data and their metadata, and encryption status. Logging activities, defining alerts, and automating workflows can be done with AWS Lambda without managing additional infrastructure.

Security is, of course, one of the most important aspects in IT environments, and AWS takes it seriously too. With S3 Block Public Access, Amazon S3 is the only object storage in the public cloud today that can block public access to all objects at the bucket or the account level. Amazon S3 is compliant with many compliance programs such as PCI-DSS for financial institutions, HIPAA and HITECH for healthcare customers, the European Union’s strict GDPR privacy regulations, and many others.

In a hybrid cloud infrastructure using NetApp Cloud Tiering, Amazon S3 becomes a dependable and low-cost capacity tier for on-prem storage users. Any data that your AFF system doesn’t need for frequent access can be stored in S3 buckets securely and at a consumable price that matches the usage pattern. Cloud Tiering supports two Amazon S3 storage classes across multiple worldwide AWS regions: Standard and Standard-Infrequent Access. Read more about the cost and performance benefits of using Amazon S3 here.

Using a Hybrid Cloud Infrastructure to Enhance the Data Center

Almost every customer underestimates how much inactive data they have stored on their primary data storage systems. After turning to NetApp Cloud Tiering, users on average find as much as 80% of the data on their AFF systems fits the profile to move to cloud-based object storage. That’s almost an entire machine’s worth of unused data! Once that space can be freed up, the AFF system’s storage space can be better utilized for additional critical and high-demand workloads instead of just housing old data.

In a way, using this kind of hybrid cloud infrastructure doesn’t make the on-prem system less important—it actually rededicates the data center to focus only on the most critical workloads.

While some customers purchase cheaper and lower-performance storage in place in the data center that they can use to manually move data from AFFs, others carry out expensive and complex rearchitecting projects for their applications to be able to access data from multiple sources in order to save money. Cloud Tiering, by leveraging NetApp’s FabricPool technology, offers a faster, less complicated way to achieve a hybrid architecture that can be configured in a few easy steps. With Cloud Tiering there is no need for any changes to be made to the application or in the data center. Plus, it can significantly lower investment in on-premises storage and lower operating costs of datacenter.

New call-to-action

Product Manager