Migrating AIX workloads to AWS: How to get started
Customers are asking us how to migrate their AIX workloads to Amazon Web Services (AWS). Customers have questions regarding assessing the current AIX portfolio, estimating the total cost of ownership (TCO) on AWS, and guidance in determining migration patterns and target state architectures for specific workloads. To accelerate your AIX migrations to the cloud you’ll need to know how to get started on this journey and the migration patterns available. AWS can assist you based on our experience, learnings, and guidance and can help minimize complexity, and risk, and provide clarity for your migration.
IBM AIX is a UNIX operating system sold by IBM. Created in 1986, it became one of the leading operating systems for IBM Power Systems and other IBM hardware platforms. Many popular software packages were created for AIX, and a substantial customer installation base still runs mission-critical workloads on AIX today. Unlike Windows and Linux workloads which can be migrated using a lift and shift migration pattern supported by AWS Application Migration Service (MGN), AIX workloads do not lend themselves to a lift and shift type of migration pattern and require replatforming or refactoring. This is due to the fact that the operating systems and processor architectures in Power AIX systems and x86 Linux systems differ. Although the data living inside these applications can be migrated between them.
In this blog, I will discuss your options in partnership with AWS, leveraging our various programs and partners, for migrating your AIX workloads to AWS. These options can include replatforming your AIX workloads for AWS, full modernization of the workload using AWS services, or architecting a hybrid solution with a partner product to reduce risk and provide an architecture to support incremental application modernization over time.
Why Migrate Off AIX
The business benefits of migrating off the AIX platform can be similar to migrating any on-premises data center application to AWS. Namely, increasing agility and resiliency, reducing costs, improving elasticity, and addressing risks around skills shortages. Increased agility enables your business to respond to change quickly, like creating new applications or features or enabling new channels. Organizations can reduce costs by moving from a CapEx to OPEX with a pay-as-you-go (PAYG) billing model. AWS improves elasticity by providing on-demand scaling of compute, storage, and network resources with the added value of improved application resiliency via the AWS global infrastructure. Finally, specifically for AIX, finding skilled staff who can administer these systems can prove difficult to attract and retain, which are a risk to your business operations.
Cloud Total Cost of Ownership (TCO) Assessment and Migration Planning for AIX
When planning your cloud migration project, a business case analysis is advisable to incorporate into your decision-making process. Estimating a total cost of ownership (TCO) is a standard method used to calculate the yearly costs for running the proposed workloads on AWS. TCO analysis are by nature not intended to be detailed implementation-specific guidance for application run time environments. To start any TCO analysis, a discovery of the source AIX environment is required as well as a mechanism to determine right-sized configurations and associated AWS-specific Amazon Elastic Compute Cloud (Amazon EC2) instance classes, types, and AWS storage options. To further prepare for the migration it is important to understand the application dependencies, existing server inter-communication, and identifying application dependency groups, all of which are essential to the migration and wave planning process. AWS has guidance, mechanisms, and partners to help our customers get started:
1. The AWS cloud economics team can assist you in conducting a thorough current state environment assessment using standard AIX utilities like lparstat and HMC Scanner for POWER report. These outputs can be used as inputs into AWS’s cloud economics assessment tools for estimating EC2 costs in AWS. As a note, these tools do take into account the processor differences between the POWER family and X86 architectures as discussed in the blog post Estimating Amazon EC2 instance needed when migrating ERP from IBM Power Systems.
2. AWS partners like Flexera, Device 42, and ModelizeIT have various levels of support for the IBM AIX operating system and can conduct asset discovery, business case analysis, right-sizing, and cost estimation of running the workloads on AWS. In addition, these partner tools can support migration planning deliverables such as application dependency mapping, server intercommunication, and wave planning. If you have a robust configuration management process and associated configuration management database (CMDB) you can export the information into .csv file for upload into the AWS Migration Portfolio Assessment tool. This is a complimentary tool developed by AWS for your partner or AWS Professional Service teams to leverage as you work with them to plan your migration to AWS.
Common AIX Workloads and Migration Patterns
You may be looking to migrate a variety of mission-critical software applications off of AIX to AWS. By migrating to AWS, you gain increased agility, elasticity, and reliability over existing on premises capabilities. If these benefits of the cloud are sufficient for you and you desire to minimize risk and take a longer-term view of modernizing your applications, acquiring equivalent Linux-based software for use on the AWS infrastructure may be the appropriate pattern. If you desire to reduce operational tasks and costs as you migrate to cloud then leveraging managed services may be a more appropriate pattern choice where you can remove operational tasks and reduce licensing costs with the option of moving to an open source solution.
We see customers with a minimum of three types of workloads that make up their AIX estate.
- Custom Java applications running on a commercial Java application server
- Applications leveraging commercial database software
- Line of business applications implemented with commercial-off-the-shelf (COTS) software
Java applications running on AIX are often not containerized and can leverage vendor-specific Java implementations that require Java licensing support costs. You may find these applications cannot respond quickly to changing customer requirements, are inflexible, and costly. If you have these applications you can consider increasing agility and flexibility by containerizing your applications and leveraging a managed container service like AWS Elastic Container Service(ECS) and reducing license costs by using a free Java implementation or open-source application server.
- AWS Corretto is a no-cost, Java SE standard compatible, production-ready distribution of the Open Java Development Kit (OpenJDK). Corretto comes with long-term support that includes performance enhancement and security fixes.
- AWS has guidance for Migrating from IBM WebSphere Application Server to Apache Tomcat on Amazon EC2 and Migrate from Oracle WebLogic to Apache Tomcat (TomEE) on Amazon ECS.
- App2Container is a command line tool that helps transform existing applications running in virtual machines into containers, without needing any code changes. A2C discovers applications running on a server, identifies dependencies, and generates relevant artifacts for seamless deployment to Amazon Elastic Container Service (Amazon ECS) and Amazon Elastic Kubernetes Service (Amazon EKS).
If you desire to break down your monolithic application into microservices and would like to start transforming your existing monolithic application, our AWS partner vFunction along with AWS Migration Hub Refactor Spaces may prove a good way to get started on that journey with an automated, scalable, and repeatable process.
Commercial database software often supports applications running on AIX. Moving away from these platforms can significantly reduce software licensing costs but can be a complex IT challenge. Challenges you may face when deciding to migrate away from these platforms can involve using proprietary database code, stored procedures to house business logic and a complex data model built over years. AWS has methodologies and financial incentive programs to assist you with either moving these environments to a managed service or full modernization of the database and refactoring the application to an open source database like AWS Aurora.
- The AWS Database Freedom program exists to assist you with these migrations by providing funding, advice and support. These migrations can entail retaining the existing commercial database and moving to a managed AWS service for the same commercial database.
- The Database Migration Accelerator (DMA) program is designed to assist you with the complex challenge of moving from a commercial database to an open source database like AWS Aurora. The program offers migration tools, workshops, guidance, and support for a proof-of-concept to assist our customers with refactoring the database and the application stack if necessary.
- AWS has services to assist you in the database migration journey including AWS Schema Conversion Tool (SCT), AWS Database Migration Services (DMS) and AWS DMS Fleet Advisor for analyzing, planning and migrating your on premises commercial databases.
Commercial off the shelf (COTS) Applications
Industry or line of business specific COTS software can be another complex migration to the cloud. Various challenges arise, including cross-platform migration from AIX to X86, application upgrade requirements and dependencies, and the migration to the cloud itself are all things you need to plan for. AWS has support programs and migration teams that specialize on these COTS workloads. They include AWS for Oracle Applications and SAP on AWS. These are two of the more popular COTS software vendors in the market but other COTS software vendors have a presence in the AWS Marketplace. Chances are your specific COTS software product has a supported roadmap for AWS adoption.
Retain and Incremental Modernization
Although, ceasing to invest in the AIX platform and applications by migrating off AIX is your stated direction, this may not be possible in the desired timeframes. It takes time to vacate a data center facility or fully rewrite the workloads to an event-based cloud-native architecture. For these reasons you may find retaining your AIX environments for a designated time period makes sense. Retaining the environments through an AWS managed services partner like Connectria can provide a solution to resolve these challenges and present opportunities:
- Vacate your existing data center and eliminate existing AIX hardware contracts
- Retain the existing AIX COTS software given vendor requirements or limitations
- Retain the existing AIX applications as an enabler for application modernization
- Retain the existing AIX applications while building new capabilities and services in AWS
Connectria has partnered with AWS to provide you the option of retaining your existing AIX applications while enabling integration with AWS through a pre-built, low-latency network. This architecture provides a managed service for your AIX environments and low-latency network integration to your AWS environment to facilitate the use of AWS to build new cloud-native applications using various data integration patterns. This type of architecture can provide you with low risk, incremental approach to application migration and modernization in AWS.
Conclusion – Call to Action
If you have AIX workloads as part of your IT landscape you have options when it comes to both migration and modernization of these applications. Understanding how to discover, assess and plan for your migration is the first step. Next, understanding the application portfolio and options for cloud transformation and the effort required for these transformations is important data for you when making these decisions. AWS has the programs, services and experience to assist and direct you when deciding how to move forward with your AIX migration. Now is the time to start to address these applications with AWS.
About the author: