AWS Launches New AI Agents To Simplify Legacy Migrations
Las Vegas — Amazon Web Services released on Tuesday three new artificial intelligence capabilities that will allow developers to more easily modernize and migrate legacy applications en masse, specifically those on Windows .NET, VMware and mainframes.
The AI capabilities are offered through Amazon Q Developer, which is Amazon’s generative artificial intelligence (AI) assistant for developers.
Typically, such migrations are a pain point for developers; and in some cases, such as mainframe migrations, they can be multiyear projects. But AI is changing that dynamic, said AWS CEO Matt Garman at this week’s AWS Re:Invent developer conference in Las Vegas.
“What happens is Q Dev launches agents that can automatically discover incompatibilities, generate a transformation, plan and refactor your source code, and it can do this across hundreds or even thousands of applications in parallel,” Garman told audiences during his keynote address. “It turns out Q Dev can help you modernize .Net applications four times faster than doing it manually.”
To start a transformation, a developer selects “Transform” under the AWS Toolkit section in their integrated development environment (IDE) and confirms the specific file they want to modernize. Amazon Q then uses AI agents to automatically identify the components that need to be upgraded, create a transformation plan, fix any build errors, and execute the plan.
The agent performs tasks such as upgrading existing code and configuration files, generating new files it needs and fixing issues identified with failed builds. It also presents a summary to the developer.
Windows to Linux Migration
More customers are seeking to migrate applications off Windows .NET to Linux, particularly in Europe, he said. Drivers for the migration include simplifying the adoption of modern development practices, reducing licensing costs, enhancing security, and optimizing performance.
Moving to Linux can result in a 40% savings in licensing costs, Garman said.
The new capability can create substantial time savings as well. He pointed to the European company Signatureit, which specializes in digital transactions. Signatureit wanted to move off Windows to Linux, but estimated it would take six to eight months to do so.
Using Q Developer enabled Signatureit to reduce that timeline to a few days, Garman said.
VMware to Cloud Migration
The AI can also be used to transform VMware workloads, which more companies want to do as they move away from on-premise data centers to the cloud.
“Many customers are actually happy for a portion of their existing VMware workloads to stay running on VMware, but they don’t want them to stay running over data centers. They’d like to migrate those to the cloud,” Garman said.
But VMware migrations have a unique challenge: They are “deeply entrenched” in organizations’ data services, Garman added.
“In this VMware environment, because it’s been there for a long time, there ends up that there’s this spaghetti mess of interconnected applications,” he said. “It’s actually really the hardest part about modernizing — finding out what are [the] dependencies of those applications.”
Last week, AWS announced the preview of Elastic VMware Service, which makes it easier to move VMware subscriptions to AWS and run the full VMware Cloud Foundation stack of natively on top of EC2.
“In this VMware environment, because it’s been there for a long time, there ends up that there’s this spaghetti mess of interconnected applications.”
— AWS CEO Matt Garman
Q Transformation for VMware applications makes it easier for companies to shift off VMware to cloud native solutions, reducing months of work down to a week, Garman explained.
“What happens is, and the biggest value here is, Q automatically identifies all your application dependencies, and it generates this migration plan for you, which really produces a ton of migration time and significantly reduces your risk,” he said. “It then also launches agents. They can convert your on-premise VMware network configurations into modern AWS equivalents.”
Q’s agents can also convert on-premises networking configurations to AWS networking equivalents in hours instead of weeks. AWS pointed to Cognizant, a global information technology services and consulting company that helps organizations modernize technology, reimagine processes, and transform experiences. Cognizant is using Amazon Q to help its customers accelerate the VMware modernization process, says AWS.
Mainframe Migrations
Mainframe migrations are particularly painful for organizations and can take three to five years.
“Just the effort of trying to analyze, document, [and] plan mainframe modernization is often too much,” Garman said. “I don’t know about you, but planning a project for three to five years is nearly impossible. A lot of times they just don’t get done.”
Part of the difficulty is that mainframe code is not very well documented, if at at all, which is a problem since a lot of mainframes run COBOL code. This is a major issue for companies, particularly in financial services, which are stuck maintaining systems that rely on a language few developers practice today.
Q can take those millions of lines of COBOL code and build documentation that tells you what it does, Garman said.
It can also help with planning and refactoring applications involved in mainframe modernization, starting with IBM z/OS mainframes, according to an AWS press release.
“Now I wish I could stand up here and tell you that I’m going to make mainframe migrations one click. We’re not quite there yet,” he said.
However, he added that early customer feedback and internal testing indicate that Q Developer can help organizations turn a multiyear effort into a multiquarter effort, cutting by more than 50% the time to migrate.
“If you can take a multiyear effort and break it down for a couple of quarters, that’s something that people really get their heads around,” he said.
Correction: Story updated Dec. 17 to correct that Q developers a migration plan.