Caylent Catalysts™
AWS Control Tower
Establish a Landing Zone tailored to your requirements through a series of interactive workshops and accelerators, creating a production-ready AWS foundation.
Learn how to migrate from VMware to Amazon EC2 and avoid VMware licensing and cost uncertainties while unlocking transformative cloud scalability and efficiency.
For companies using VMware or VMware on AWS Cloud (VMC), migrating from VMware products to AWS native is becoming a strategic priority. For some customers the driver is uncertainty with upcoming license changes and potential price increases, for others it’s the desire to take advantage of more cloud-native technologies. Moving to AWS native services can help in both cases.
In this blog post, we will look at the different strategies that you can use to migrate from a VMware environment to an AWS Native environment, taking advantage of all that AWS has to offer.
Preparation is the largest factor in success when it comes to any migration. Your migration should start with a discovery and planning phase, where you identify all the resources that need to be migrated, the supporting infrastructure and technologies to make the migration successful, and a plan for the migration in waves. Technical staff with years of experience in VMware features and functionality will need training, enablement, and hands-on experience to maintain or exceed the level of service the organization is used to from the legacy environment. Caylent combines workshops on feature equivalents with a co-delivery model that allows teams to gain practical experience in AWS while being supported by experts.
It’s important to start by considering things like authentication and networking that need to be set up in the target environment before you can start migrating servers. These technologies will already be in place in your source environment and equivalents need to be created in the destination environment early to ensure success. Ensure that the teams migrating the VMs will have the required levels of access and that the target accounts have the right security controls in place. If your target environment is a new AWS environment, consider building out an AWS Control Tower implementation or taking advantage of Caylent's Control Tower Catalyst to help prepare the environment for receiving your inspiration.
For relatively small environments, it may be possible to migrate everything in one wave. For most migrations, you are going to have to plan migrations in waves. Each wave should identify a selection of target applications, and their supporting infrastructure. You should also identify allowable windows for down time as the applications are cut over.
For each migration wave, select an appropriate strategy for the migration. The most common strategies include:
Redeployment:
This is the easiest strategy. Applications are redeployed onto new EC2 instances. Data is typically migrated via a backup and restore. This is a good candidate for applications that can easily support downtime windows. Applications that target internal users, or applications that are in use during business hours only as good candidates for this type of migration.
Replication:
AWS provides tools such as the Application Migration Service (MGN) that allow you to replicate a running server via block-level replication. You can then perform a final cutover once an application has been completely migrated. Replication is a good candidate for applications that are in constant use, and can not easily support a maintenance window long enough to support a redeployment strategy. For many workloads, leveraging a replication migration strategy will be preferable as it limits downtime without a huge increase in cost or complexity.
Zero Downtime Migration:
Zero down time, or near zero downtime migrations, typically start with replicating the application data and databases from the source environment to the destination. For this ongoing replication, customers will normally leverage database native tools, such as SQL Server Always On asynchronous replication. Tools like AWS DataSync can be used to synchronize data files, if required. Once the application data has been synchronized to the target system, you can deploy any user application in the AWS native environment. Application cutover is then performed during a maintenance window by rerouting the DNS entries to point at the new copy of the application. While the target is to perform a zero downtime migration, you should always plan to perform the cutover around a maintenance window just in case something goes wrong and you need to revert DNS entries back to the original implementation.
Zero-Downtime migrations should be planned for only mission-critical applications that are under constant use, and can not tolerate downtime. Successfully performing zero-downtime migrations of data requires extensive planning, testing, and a high level of application-specific knowledge to ensure that the team understands all of the nuances of the application.
There are a lot of options and significant changes to plan for. Caylent can help navigate the maze of alternatives with our VMware to AWS Strategy offering.
Customers who migrate from VMware to Amazon EC2 can gain access to a range of benefits, including additional service offerings, optimizations, and cost controls. AWS has a catalog of over 200 fully featured services that are easily accessible from EC2 instances. Because these services exist in your account, interacting with them is both easy and highly secure.
VMC leverages EC2 instances as the backing compute for all of your workloads. These instances are AWS’s I family of EC2 instances and are optimized for high IO workloads. This choice is made because VMC leverages instance storage for the VMs. By moving from VMs on VMC hosted hardware, customers can take advantage of EC2 instances that are optimized for the specific needs of their workloads on an instance-by-instance basis.
For compute-heavy workloads, customers can take advantage of compute-optimized EC2 instances. For memory-intensive applications, such as SQL Server, customers can leverage AWS families of memory-optimized instances. For IO-intensive applications, customers always have the opportunity to leverage IO-optimized instances with on-instance NVMe storage. Additionally, EC2 provides customers with access to a range of storage options to address specific needs. Learn more about instance types here.
By leveraging EC2 instances directly, customers can match their utilization to their needs. There is no need to buy an entire additional VMC host in order to support a few new virtual machines. Additionally with EC2s per second billing, it is easy for customers to spin down instances that are no longer needed in order to save money.
Migration to the cloud can be a challenge for businesses that have never performed a migration in the past. If you have migrated data centers, then you will understand some of the complexities involved, but cloud can have its own unique nuances that you need to be aware of.
Plan your migration
The biggest mistake you can make is just jumping into a migration without planning. Make sure you define migration waves that move applications as groups. Make sure you understand all of the servers that are part of each group. You don’t want to migrate an application server to EC2 and leave the database behind. You don’t want to have an application that is spread access to your AWS Account and legacy environment. A big part of planning your migration is to really understand the applications that you are migrating and how business-critical the applications are. What can you afford for downtime?
Understand your service level agreements
While zero-downtime migrations sound like they should be the default for all applications, this kind of migration adds substantial complexity and effort due to the ongoing replication of data. Additionally, zero downtime migrations may be impossible depending on the databases being used. Understanding your service level agreements is critical to understanding the requirements for an application. Choose the simplest migration path that will work for the specific application.
Start with the foundation
You can’t start a migration until you have the underpinning technologies in place. This includes things like networking and Authentication / Authorization. These foundations are difficult to change once a migration is in place. Invest in getting these elements in place first. If you don’t already have a foundational environment set up in AWS, look at AWS Control Tower or the Caylent Control Tower Catalyst.
Don’t treat the cloud like a data center
Simply moving all of your infrastructure to EC2 instances and treating them like servers in the data center won’t allow you to realize all of the cloud's benefits. Unfortunately, this is exactly what a naive VMware migration leads to. As you shift away from VMware and into EC2, you should plan to take advantage of the unique features that the cloud offers. Cloud-native technologies like Auto-Scaling can help bring resilience to workloads. Cloud-native backups and cross-region replication may help reduce your dependency on licensed software.
Take workload requirements into account when planning
Each workload, and potentially each instance, will have unique characteristics. Get to know your server requirements, and choose instances that are right for those requirements.. If you aren’t sure of what kind of instance makes sense, start with the M family of instances and monitor performance. You can change your instance type, in most cases, at the cost of a reboot.
Right size your instances
The first step in controlling costs in the cloud is to right-size your instances. In almost every case, each increase in the size of an EC2 instance doubles the amount of vCPUs and memory but also doubles the cost. This means that any instance you can move down one size, will result in a 50% savings on the compute portion. (Note the storage costs will remain the same.) When helping customers to prepare to migrate, we often encounter servers with very low (< 25%) utilization for memory and / or CPU. Right-sizing those down will save you money. Similarly, once you have completed a migration, monitor the performance of your applications. You may find that running on EC2 natively provides you with opportunities to further reduce instance sizes. This is especially true if your instances are running memory or CPU bound instances. Right size before and after your migration to get the best value out of EC2.
All storage is not equal
If you come from the data center VMware world, or you are sitting in VMC, you probably have file servers that are being used to store terabytes of files. In most cases, these file servers do not require high-performance characteristics. Take advantage of options like FSx for Windows File Server to store all of those files rather than spinning up an EC2 instance that you have to manage. Match the storage characteristics to your actual needs. For example using FSx for Windows Server, with magnetic drives, can reduce your storage costs to as little as $0.0065 per gig per month for files that can afford lower performance and a slightly lower SLA on availability.
One size, or strategy, does not fit all
Choose options and strategies that are right for the workload in question. You don’t have to migrate every workload the same way, and you don’t have to land everything on the same technology stack. Success comes from understanding that migrating to AWS Native creates new options.
Leverage AWS Partners
Partners can dramatically accelerate your migration from VMware to AWS. While your staff have existing day jobs that will impact a migration you execute yourself, a partner is going to be dedicated to making your migration successful. Partners like Caylent have the experience of doing migrations for many customers and have seen a multitude of different environments.
Migration from VMware to AWS can seem daunting. However, successful migration planning, including migration waves and getting the foundational elements in place ahead of a migration, can result in a straightforward and successful migration to AWS native services. Caylent has helped thousands of customers make successful migrations to AWS including migration from datacenters, other cloud providers and VMC. Explore our VMware to AWS Strategy offering or get in touch with our experts to find out how we can help.
Mark Olson, Caylent's Portfolio CTO, is passionate about helping clients transform and leverage AWS services to accelerate their objectives. He applies curiosity and a systems thinking mindset to find the optimal balance among technical and business requirements and constraints. His 20+ years of experience spans team leadership, technical sales, consulting, product development, cloud adoption, cloud native development, and enterprise-wide as well as line of business solution architecture and software development from Fortune 500s to startups. He recharges outdoors - you might find him and his wife climbing a rock, backpacking, hiking, or riding a bike up a road or down a mountain.
View Mark's articlesCaylent Catalysts™
Establish a Landing Zone tailored to your requirements through a series of interactive workshops and accelerators, creating a production-ready AWS foundation.
Caylent Catalysts™
Reduce cloud cost and migration complexities when moving from VMware to AWS.
Explore how we helped our customer in the financial sector migrate from High-Performance Computing (HPC) workloads on an on-premise Slurm cluster to AWS ParallelCluster, detailing the process, challenges, and benefits.
SQL Polyglot, our new groundbreaking AI-powered solution, significantly accelerates and simplifies complex database migrations, helping you minimize technical debt. Discover how it can reduce your migration time and costs by automating the translation of stored procedures.
Learn how to optimize website performance by converting images to WebP format using Amazon CloudFront and S3, improving load times and user experience.