Amazon FSx for NetApp ONTAP is a native, fully managed AWS storage service with the full feature set of a NetApp® ONTAP® enterprise storage system—features such as multiprotocol support, NetApp Snapshot™ technology, backup, cloning, and replication. It has become the platform of choice for customers seeking to easily launch, run, and scale applications on AWS.
This blog illustrates the most common challenges and use cases that we see when working with customers who are extending or migrating workloads into AWS. We’ll also present a best practice blueprint for assessing, migrating, and protecting enterprise applications with FSx for ONTAP.
The need for a comprehensive hybrid cloud migration strategy
To continue digital transformation efforts, an ever-increasing number of organizations are modernizing data centers and looking for ways to move workloads to AWS.
Key objectives include:
- Modernizing data centers
- Reducing the number of data centers to manage
- Updating data protection strategies using a 3-2-1 backup strategy
- Using automation for ITOps
- Optimizing cloud costs
Many of these organizations have a mix of enterprise database workloads, general-purpose file shares, and virtualized workloads running on different storage arrays. Some data has been sitting idle in storage for so long that it’s virtually impossible to identify ownership and determine whether the content should be migrated. For all successful migration into AWS, we’ve identified five necessary steps to complete the journey:
- Step 1: Assessment
- Step 2: Discovery
- Step 3: Migration
- Step 4: Protection
- Step 5: Optimization
Step 1: Assessment
Assessment is a critical first step before any AWS migration to ensure that your applications’ requirements for performance, cost, compliance, and security can be delivered. NetApp offers several tools to help you understand your workloads and identify your migration strategy.
Evaluating workloads
NetApp Cloud Insights can identify all the resources to be moved and map them to analogous resources in FSx for ONTAP. Cloud Insights is able to quickly discover almost everything in the environment across data center and cloud boundaries, including virtual machines and storage volumes. Its flexible filters and finders help you search for assets at a granular level based on a wide range of criteria.
If you want your migration to be as cost effective as possible, you must rightsize resources as you migrate them. As shown in Figure 1, Cloud Insights also provides recommendations for the cloud compute and storage resources needed for each workload, including memory, throughput, and I/O requirements.
Figure 1) NetApp Cloud Insights rightsizing storage resources and estimated savings.
Existing NetApp customers can also use the NetApp Active IQ® Digital Advisor tool to identify application usage. In the example in Figures 2 and 3, Active IQ identifies six workloads to migrate to AWS:
- File shares
- Oracle
- SAP
- Microsoft SQL Server
- VMware
- SharePoint
FSx for ONTAP is a prime solution for each of these workloads, because it offers a fully managed ONTAP environment in AWS that can grow (and shrink) automatically as business needs dictate. FSx for ONTAP uses NetApp storage efficiency technologies to control storage costs by optimizing resource utilization and minimizing the amount of data stored. This results in a 3-year return on investment of as much as 61% with investment payback in under 6 months. It also costs, on average, 31% less than self-managed, on-premises storage. Read the full Forrester “Total Economic Impact of Amazon FSx for NetApp ONTAP” study.
Figure 2) In this example, Active IQ Digital Advisor identified six workloads to migrate to AWS.
Figure 3) A deeper dive into the six workloads.
Preparing data for migration
NetApp BlueXP classification service provides discovery, mapping, and classification of unstructured and structured data to give organizations visibility into their entire data estate, whether it’s in AWS, on premises, or both. Using AI-driven analysis, BlueXP classification provides critical information, such as type, size, time attributes, ownership, and user/group permissions. It also analyzes data to identify personal identifiable information (PII), such as emails, credit card numbers, international bank account numbers (IBANs), national IDs, IP addresses, passwords, ethnicities, religious beliefs, and more. It then classifies the PII into sensitivity levels (standard, personal, sensitive-personal) and categories (HR, legal, marketing, sales, finance, and so on).
By using BlueXP classification before migrating to the cloud, you can reduce costs and accelerate your migration by removing duplicate, stale, or unneeded data after cloning the storage volume. The cloned volume is then used as the source for migration. BlueXP classification can also use data context to identify and secure sensitive data.
Figure 4) NetApp BlueXP classification savings opportunities and data classification.
Step 2: Discovery
NetApp BlueXP is the unified control plane for discovering on-premises ONTAP clusters and any existing AWS cloud environments, such as FSx for ONTAP, NetApp Cloud Volumes ONTAP, and Amazon S3. By bringing each of these environments into BlueXP, you can manage volumes and perform advanced data management—such as data replication, backup and recovery, data classification, and data tiering—across your entire hybrid cloud environment without ever leaving the BlueXP canvas.
BlueXP combines data protection through block-level, incremental-forever backup (including replication, mirroring, and synchronization options), with advanced ransomware protection to protect the entire data estate.
Step 3: Migrating to FSx for ONTAP
Migrating from NetApp ONTAP
Rehosting, also known as “lift and shift,” is one of the easiest, fastest, and least-expensive ways to migrate an existing workload into AWS. FSx for ONTAP makes this process easy because of its seamless integration with ONTAP-native workloads and flexible support for block and file workloads. You can lift and shift applications and workloads from on-premises storage into AWS without making significant architecture changes.
You can easily migrate file shares that reside on on-premises ONTAP systems to FSx for ONTAP by using NetApp SnapMirror® technology. Even for volumes with complex directory structures, over 50 million files, and very small file sizes (on the order of kilobytes), SnapMirror employs block-level replication between two ONTAP file systems to make it quick and efficient. During the migration, deduplicated and compressed data remains deduplicated and compressed, which reduces transfer times and the amount of bandwidth required. Snapshot copies on the source ONTAP volumes are preserved when migrated to the destination volumes. Figure 5 shows replication times for initial copy and incremental data changes with SnapMirror by capacity and network bandwidth.
Figure 5) Replication times for initial copy and incremental data changes with SnapMirror by capacity and network bandwidth.
Migrating your on-premises ONTAP file systems to FSx for ONTAP involves the following high-level tasks:
- Create the destination volume in FSx for ONTAP.
- Gather source and destination logical interfaces (LIFs).
- Establish cluster peering between the source and destination file systems.
- Create a storage virtual machine (SVM) peering relationship.
- Create the SnapMirror relationship.
- Maintain an updated destination cluster.
- Cut over to your FSx for ONTAP file system.
For detailed information about any of these steps, check out the guide to migrating to FSx for ONTAP using SnapMirror in the AWS documentation library.
Figure 6) Typical enterprise workload migration architecture.
Migrating from third-party storage
To migrate data from different storage solutions to FSx for ONTAP, AWS DataSync is the tool for the task. AWS DataSync makes it easy to transfer data between on-premises volume exports/shares and FSx for ONTAP. It requires minimal setup, and you pay only for what you use.
Transferring files from a source to a destination using DataSync involves the following basic steps:
- Download and deploy the DataSync agent in your on-premises environment and activate it.
- Create a source and destination location.
- Create and run the transfer task.
The AWS documentation library has more information about migrating to FSx for ONTAP by using AWS DataSync and a comprehensive AWS DataSync user guide. For techniques to improve planning and maintain sufficient capacity during data migration, check out this blog: Migrating file shares to Amazon FSx for NetApp ONTAP using AWS DataSync.
Step 4: Protecting on-premises and AWS workloads with a 3-2-1 data protection strategy
Although there are multiple approaches that an organization might take for data protection, the 3-2-1 data protection strategy has become an industry standard. With this strategy, you have three copies of data in two formats (for example, NFS and object), one of which is off site from where the backup was taken. A 3-2-1 strategy doesn’t guarantee that the data will never be compromised; rather, it’s sufficient for recovering from virtually any failure.
Figure 7) Typical 3-2-1 data protection architecture.
3-2-1 for on-premises workloads
For workloads that are on premises running on NetApp storage, you can use BlueXP to set up the following relationships:
- Three data copies:
- One production copy on premises
- One disaster recovery copy to FSx for ONTAP using SnapMirror from on-premises production storage
- One archive copy to Amazon S3 using NetApp Cloud Backup from on-premises production storage
- Two formats:
- ONTAP flash on premises and in the AWS Cloud
- Object storage in Amazon S3
- One off-site copy:
- FSx for ONTAP or Amazon S3
Figure 8 ) Dragging on-premises environment onto instances of FSx for ONTAP and Amazon S3 for disaster recovery, backup, and archive purposes.
3-2-1 for workloads in AWS
For the workloads that have migrated to FSx for ONTAP, setting up a 3-2-1 backup strategy is easy when you use native tools in AWS or drag-and-drop tools in NetApp BlueXP:
- Three data copies:
- One production copy in FSx for ONTAP
- One disaster recovery copy to another FSx for ONTAP instance in a second region by using SnapMirror cross-region replication
- One archive copy to Amazon S3 by using AWS Backup from FSx for ONTAP production
- Two formats:
- ONTAP flash in FSx for ONTAP
- Object storage in Amazon S3
- One off-site copy:
- FSx for ONTAP instances in a second region or Amazon S3
Figure 9) Drag an FSx for ONTAP environment onto another instance of FSx for ONTAP for disaster recovery purposes; apply the automated backup copies by using AWS Backup from the production and disaster recovery FSx for ONTAP instances. The AWS Backup relationships (shown as green lines) do not appear in the BlueXP Canvas and are shown for illustrative purposes only.
You can then use BlueXP to modify mirroring and backup policies to fit your recovery time objective (RTO) and recovery point objective (RPO) requirements.
Figure 10) Setting protection policies.
Figure 11) Setting up data replication schedule policies.
Technical resources are available to help you as you replicate and protect critical applications and workloads such as SAP HANA, Microsoft SQL Server, and VMware, including the following blogs:
- VMware Cloud on AWS Disaster Recovery with Amazon FSx for NetApp ONTAP and SnapMirror
- Protect your SQL Server workloads using NetApp SnapCenter® with Amazon FSx for NetApp ONTAP
- Replicate and clone SAP HANA databases with Amazon FSx for NetApp ONTAP
Step 5: Optimization
After data is migrated and production has cut over to AWS running on FSx for ONTAP, it’s time to monitor and optimize the environment. This is where we recommend using NetApp Cloud Insights to monitor resource utilization. Cloud Insights can monitor things like:
- CPU utilization
- Memory utilization
- Storage utilization
- IOPS
- Throughput
Cloud Insights provides performance and usage data that helps you rightsize AWS resources so that you aren’t paying for more performance or capacity than you need.
Figure 12) Cloud Insights dashboard showing total compute resources in your AWS environment.
Figure 13) Cloud Insights dashboard showing virtual reclamation and remediation.
When you use FSx for ONTAP, it can automatically tier cold or inactive data from high-performance SSD storage to capacity pool storage to optimize costs and utilization. There are four tiering policies you can choose from:
- Auto: Moves all cold data to the capacity pool tier as determined by the defined cooling period (2 to 183 days).
- Snapshot: Moves only Snapshot data to the capacity pool storage tier.
- All: Marks all user data and Snapshot data as cold.
- None: Keeps all data on the primary storage tier and prevents it from being moved to capacity storage.
Conclusion
Figure 14 shows the BlueXP canvas after you’ve migrated the workloads to FSx for ONTAP and set up a 3-2-1 data protection strategy. It provides a visual representation of the backup, replication, and copy-and-sync relationships between your on-premises and cloud storage environments.
Figure 14) BlueXP canvas depicting the replication and backup relationships between the on-premises and AWS storage environments. The AWS Backup relationships to Amazon S3 (shown as grey lines) do not appear in the BlueXP Canvas and are shown for illustrative purposes only.
In summary, using Amazon FSx for NetApp ONTAP for migration and data protection offers the following benefits:
- Cost savings:
- Data compression automatically compresses data blocks, reducing the amount of storage space consumed—saving up to 87%, depending on the application.
- Data deduplication eliminates duplicate data blocks by saving a pointer to the original block, reducing storage usage by as much as 90%, depending on the application.
- Data compaction automatically combines remaining small chunks of data into single 4KB physical blocks.
- Thin provisioning allocates storage only when data is being written to a volume rather than allocating ahead of time based on peak demand.
- Recovery point objective as low as 5 minutes.
- Recovery time objective in minutes with automatic failover to FSx for ONTAP.
- Improved cyber resilience with multiple data types—no single point of failure—with 3-2-1 backup.
- Storage-efficient data protection capabilities create additional savings:
- NetApp Snapshot technology creates point-in-time Snapshot copies that are extremely space efficient, saving only data changes.
- Using Snapshot technology, NetApp FlexClone® writable data clones are created instantaneously, consuming storage only for changes that are made to the clone.
This blog has presented a high-level overview of migrating workloads into AWS. There are many other specific areas you might need to pay attention to, such as workload separation and billing. If you have any questions, feel free to contact your account team at AWS and NetApp, and we’ll be glad to meet with you.
Learn more about Amazon FSx for NetApp ONTAP.
