The hype around Artificial Intelligence (AI) and machine learning is at a staggering, all-time high. At the same time, the reality is that in order to be ready to take advantage of these highly sophisticated capabilities you have to first look to modernization approaches for your operations, infrastructure, and ultimately your application source code and data. The existence of technical debt as a gating factor towards cloud adoption is still vastly under-represented by the technology industry at large.
It was refreshing to see Jason Zander, Executive Vice President of Microsoft Azure lend airtime to application modernization during Tuesday’s mainstage performance, although the focus seemed more about the destination rather than the journey. Later that day there was a breakout session focused on rehosting. While it was great to see this as a topic of discussion, rehosting is simply one option among many different possible mainframe application modernization disposition strategies. To take advantage of emerging cloud capabilities and move beyond simply “lifting and shifting” workloads to the cloud, customers need to think about selecting a combination of modernization strategies, including automated conversion, refactoring and re-engineering.
Unfortunately, there still seems to be a fairly large gap in the amount of awareness being allocated to the realities of what state your business and IT needs to be in before you can effectively take advantage of emerging technologies.
The effort required to move applications from legacy platforms can be daunting, and you need a proven and trusted partner to do the heavy lifting. Not enough attention or practical education is being given to this critical and fundamental aspect of transformation as it relates to discussions about the journey to the cloud, AI and machine learning.
The term “legacy” when it comes to operations, infrastructure and applications is still poorly defined by the industry at large. As we walked the show floor and attended networking events, we asked people how they define the term and we came up with no less than 5 different definitions.
From our perspective, mainframe is a different starting point to other “legacy” modernization projects. Legacy mainframe modernization projects involve decades of complex integrated applications developed in different languages and accessing different data stores – all supported by a complex operational and infrastructure landscape requiring unique yet highly diminishing skills.
A true legacy modernization project must begin with an assessment and design phase to define the conversion scope and approach, validate the desired disposition strategy, and document a clear plan to deliver. A thorough assessment can reduce scope by removing unused, unreferenced objects and can save your organization around 40-70 percent right out of the gate. A long term strategy we see with our customers is the desire to move to a cloud-native environment supported by microservices. In this case, a good assessment will also leverage top-down approaches and business users to determine what makes sense to become a microservice, and what can stay a monolith. Monolithic applications must first be untangled before they are strangled, to modify a phrase from the godfather of microservices, Martin Fowler.
We all want to realize the benefits of fully embracing elastic cloud growth models and emerging technologies that are reshaping the fabric of IT. As an industry we first need to properly educate customers about the risks they face if they don’t modernize their underlying technologies. IT modernization is a journey, and you won’t be able to progress in emerging areas like AI and machine learning if you don’t take a hard look at your legacy IT as a first step in that journey.
We enjoyed seeing many of you in Las Vegas this year. If we didn’t get to see you, we look forward to meeting up at Microsoft Ignite Orlando 2018 in September.