Given the current market conditions, migrating to Amazon Web Services (AWS) is a cost-effective option for enterprises deeply invested in VMware vSphere. The reasons for VMware users to migrate workloads to the AWS Cloud in a way that allows for full flexibility and scalability are stronger than ever.
Whether the reason for migrating is to reduce costs, modernize an IT infrastructure, or enhance capabilities, AWS provides numerous services that can make it easier to carry out and operate.
In this post, we introduce you to a method of migrating VMware workloads to AWS that takes advantage of the compute power of Amazon Elastic Compute Cloud (Amazon EC2) and the fully managed file storage service provided by solid-state drive (SSD)-backed Amazon FSx for NetApp ONTAP (FSx for ONTAP).
Read on as we cover:
Reasons for migrating VMware workloads to AWS
VMware and the software-defined data center (SDDC) architecture have been a mainstay of enterprise information technology (IT) operations. The rise of the cloud has presented VMware users with a new opportunity: extending the SDDC to AWS.
Recent developments have increased interest in the resources AWS provides, specifically, Amazon EC2 and scalable storage services, such as FSx for ONTAP. These resources offer compatibility, consistency, and an easy avenue for a lift-and-shift migration from on-premises VMware workloads to take full advantage of AWS.
There are several use cases for making this transition:
- Extend your data center. Moving to AWS enables flexible capacity scaling in response to changing business needs.
- Disaster recovery. AWS can be used as a backup and disaster recovery (DR) site in a different Region if a full migration to the cloud isn’t part of a company’s journey yet.
- Shrinking or evacuating your data center. Shifting to AWS can help shrink your on-premises footprint, whether it’s to facilitate a cloud-first strategy or decommission old data centers.
- Creating new environments for developing new applications. Migrating data can help make it easier to build, test, and run modern apps, so on-premises data can be used in the cutting-edge services provided by the AWS Cloud.
VMware virtual machine (VM) migration to AWS: Solution requirements
There are numerous challenges that VMware customers look to solve in their migrations:
- Enhance enterprise data management capabilities
Moving to AWS means finding solutions for critical functions that were commonly used in on-premises environments, including efficient block-level data replication, backup and recovery solutions, and cost-reducing storage efficiency features.
- Optimize storage deployments
Some VMware users have AWS deployments that can be optimized by migrating that data to other AWS services, such as Amazon EC2 for compute and FSx for ONTAP for storage. Benefits range from attaining methods for more flexible resource allocation to cost-optimizing features.
- Reduce Amazon EC2 costs
There’s a full range of Amazon EC2 instances, and one size does not fit all. However, block storage options use bandwidth that’s tied to specific Amazon EC2 instance types. Those instance types can be costly to size appropriately to meet enterprise-grade targets for input/output operations per second (IOPS) and throughput.
Amazon EC2 instances using FSx for ONTAP offer an ideal solution for VMware users who are ready to migrate to the AWS Cloud.
Migrating VMware to AWS using Amazon EC2 and FSx for ONTAP: Overview
The migration of on-premises VMware vSphere VMs to AWS can be easy to carry out while still continuing to deliver all of the enterprise data management capabilities, storage optimizations, and cost reductions that VMware users require.
This solution uses Amazon EC2 instances for compute with data disks placed on FSx for ONTAP internet small computer system interface (iSCSI) logical unit numbers (LUNs). The migration itself is carried out using the MigrateOps, which automates all the steps involved, including provisioning mapped iSCSI LUNs, making the migration a simple declarative operation.
Let’s take a closer look at the components involved.
Amazon FSx for NetApp ONTAP
AWS and NetApp® partnered to create Amazon FSx for NetApp ONTAP, a fully managed AWS service with all the signature features that NetApp users depend on in ONTAP®-based file systems. Now, the enterprise-grade features that ONTAP is known for—high performance, data protection, storage efficiency, and multiprotocol access support for network file system (NFS) and server message block (SMB)—are built-in parts of AWS.
The combination of Amazon EC2 instances and FSx for ONTAP serving block storage and iSCSI protocol is already widely used in enterprise database deployments of MS SQL Server, Oracle, and SAP HANA databases—and it’s a deployment model that’s especially relevant for VMware users right now. FSx for ONTAP serves as a supplemental datastore for VMware vSphere deployments and can be deployed with VMware Cloud on AWS, with Amazon EC2 instances, and more. This post will focus on its interoperability with Amazon EC2.
On a technical level, there are three major components to be aware of when considering how FSx for ONTAP works with Amazon EC2:
- File System
This is the main resource FSx for ONTAP provides in the migration and ongoing operations on AWS: A highly available file system based on SSD storage.
- Storage Virtual Machines
Data in the FSx for ONTAP file system is accessed using one or more storage virtual machines (SVMs). An SVM is an isolated file server with unique access points that Amazon EC2 instance interfaces with directly to access the iSCSI LUNs hosted by FSx for ONTAP.
- Volumes
These serve as virtual containers that can use single or multiple LUNs. Data in the volume consumes physical storage capacity.
Data transfer solutions
There are several ways to get your data to AWS for use with Amazon EC2 and FSx for ONTAP. In this post, we discuss using Cirrus Migrate Cloud and the MigrateOps feature.
- The MigrateOps feature of Cirrus Migrate Cloud
Cirrus Migrate Cloud is a software-as-a-service (SaaS) offering from Cirrus Data Solutions that is available via the AWS Marketplace. MigrateOps is a feature of CMC that provides “data-mobility-as-code”: Users can manage data mobility operations declaratively via YAML operations.
With this code-based migration approach, MigrateOps streamlines the migration effort, letting it operate at scale and non-disruptively. It provides automated options for remediating the OS, application cutovers, migrating clusters, integrating virtualization platforms, integrating with storage management services, and configuring Storage Area Network (SAN) / iSCSI protocol use, among other tasks.
These can take major effort to manually configure in a migration—a burden that’s replaced by easily specifying your preferred parameters in YAML.
- SnapMirror
SnapMirror® is an ONTAP-based data replication technology that can migrate on-premises ONTAP data to AWS. SnapMirror makes data syncs quick and efficient by only transferring updated data that has been reduced with the help of ONTAP storage efficiencies. This not only reduces storage and transfer costs, but reduces the total amount of time it takes to transfer that data.
SnapMirror doesn’t come into play in the main migration method that we describe below, but there is a use case for it when migrating VMware data that will be briefly covered.
High-level architecture
This high-level architecture diagram depicts how the Virtual Machine Disk (VMDK) data, including operating system (OS) and application data, can be migrated using CMC MigrateOPs from VMware to AWS and FSx for ONTAP:
In another use case, where the application data source is an on-premises ONTAP system, there is a similar approach. The migration of the in-guest VMware LUNs to FSx for ONTAP uses SnapMirror, while the OS data is transferred to AWS using CMC, as shown following:
The benefits of deploying VMware VMs on AWS with Amazon EC2 instances and FSx for ONTAP
Users stand to gain some major benefits once their data has migrated from VMware to AWS with Amazon EC2 and FSx for ONTAP, such as:
- High throughput and low latency
Consistently high performance that VMware-based workloads are likely to demand.
- Intelligent nonvolatile memory express (NVMe) caching
Enables data to be brought closer to users so there are fewer performance issues when working remotely or between disparate locations.
- Tunable settings
FSx for ONTAP allows throughput, IOPs, and capacity to be changed on the fly, making it adaptable and responsive to demands.
- Block-based data replication
Helps keep the costs and size of data transfers to and between repositories on AWS optimized.
- Multiprotocol accessibility
Provides a single storage service capable of protocol access for workloads on macOS, Linux, and Windows.
- Highly efficient data replication
SnapMirror helps improve recovery point objectives (RPOs) by facilitating backup and DR data copies that take just minutes to complete instead of hours.
- Multi-AZ high availability
Two nodes set up across different AWS Availability Zones (AZ) preserve operability—which is especially important for specific scenarios such as Microsoft SQL Server Failover Cluster Instances (FCI) and SAP shared storage.
- Data protection
NetApp Snapshot™ technology and DR orchestrated by SnapMirror prevent data loss and speed up recovery.
- Storage efficiency features
Reduce costs and the total size of the data stored on AWS.
- Granular backup and restore
NetApp SnapCenter® unified data protection works with FSx for ONTAP to let you restore just the data you need, which speeds up recovery.
How to migrate VMware VMs to AWS using Amazon EC2 and FSx for ONTAP
Ready to get started? For a complete walkthrough on how to get your VMware data to AWS using the MigrateOps feature of Cirrus Migrate Cloud, please see this technical report: Migrate VMware VMs to AWS using Amazon EC2 and Amazon FSx for NetApp ONTAP iSCSI.