Why AI is giving mainframe infrastructure a new lease on life
By Michael Vincetic (pictured), Kyndryl’s Practice Lead for Cloud, Core Enterprise & zCloud ANZ
Despite the popularity of the cloud-first approach among enterprises, mainframe infrastructure continues to hold a place of importance for businesses, with one in four Australian enterprises running 60-70% of their organisation’s mission-critical applications on the mainframe. But the advent of Artificial Intelligence (AI) is transforming the mainframe, making it fit for the future.
Tech debt has perhaps been one of the most persistent challenges in today’s digital business environment, and mainframe infrastructure is perhaps one of the more prominent contributors to such tech debt. As a result, IT teams are under pressure to transform their digital environments and derive greater value from their mainframes.
Although the mainframe has become a cornerstone of hybrid IT strategies that have the power to support the most critical workloads, enterprises are looking for ways to modernise their mainframes to better suit new and emerging use cases while also meeting the efficiency and regulatory compliance needs of business today.
Organisations need to harness the power of AI to modernise the mainframe
When it comes to modernising the mainframe, there is an emerging trend among enterprises to tap into AI and generative AI (GenAI) developments to make the transformation process more effective and efficient. Indeed, 86% of business and IT leaders in Australia and around the world are quickly adopting AI and GenAI to accelerate their mainframe modernisation initiatives.
This is according to Kyndryl’s second annual State of Mainframe Modernisation Survey Report, which reveals that 71% of IT leaders are already implementing GenAI-driven insights as part of their mainframe modernisation strategy. Moreover, nearly half of those surveyed aim to use generative AI to unlock and transform critical mainframe data into actionable insights.
The potential that AI-enabled mainframe modernisation strategies offer for enterprises is massive. However, 80% of those surveyed in Kyndryl’s research are only in the early- or mid-stages of AI integration. This means that there is still enormous value that AI and GenAI can bring to mainframe environments and how enterprises use them.
AI-powered platforms are the key to improving operational efficiency
Operational efficiency is one of the top areas where AI has the potential to transform and bring additional value to the mainframe. For instance, AI and GenAI tools can transform the mainframe environment by delivering insights into complex unstructured data, augmenting human action with advances in efficiency, speed and error reduction.
AI-powered open integration platforms such as Kyndryl Bridge can help automate and optimise mainframe operations. They do this because AI-driven operational insights can enable more proactive and predictive management of mainframe systems, while providing visibility and control over mainframe performance and costs.
The observability such AI-powered platforms enable, especially across hybrid IT environments that involve multiple networks, systems and infrastructures, can offer the real-time insights and expanded control needed to increase productivity and efficiency, ultimately leading to faster transformation and better business outcomes.
IT leaders use AI to drive mission-critical data insights
Data insights are an incredibly important element in unlocking mainframe value, so it should come as little surprise that 44% of the IT leaders surveyed in Kyndryl’s research said they use AI to unlock their mission-critical data and transform unstructured data into accurate, unbiased, explainable and actionable insights.
Moreover, a full third of IT leaders are using generative AI to uncover business insights from mainframe-managed data to aid the development of new products or services. Enterprises can drive such insights by running AI large language models (LLMs) across mainframe data, be it in cloud infrastructure as part of a hybrid environment or the mainframe itself.
These AI-driven operational insights can support solutions that enable more proactive and predictive management of mainframe systems, even in a hybrid environment, and provide visibility and control over mainframe performance and costs — a key element in the mainframe modernisation process.
AI skills shortage can be alleviated by itself
Regardless of the forces driving mainframe modernisation, many enterprises are grappling with a skills shortage preventing them from going full steam ahead with such projects. According to Kyndryl’s research, skills are especially short in new areas such as generative AI that can, ironically, facilitate mainframe transformation and help alleviate the skills gap itself.
Additionally, security skills are in particularly high demand, no doubt driven by increasing regulatory compliance requirements that weigh heavily on businesses’ security operations, with almost all survey respondents flagging security as the key factor driving modernisation decisions. Meanwhile, 53% of people entering the workforce today lack mainframe skills.
But herein lies an opportunity. There are ways to use new and emerging AI and GenAI tools to augment existing skills without necessarily needing those skills to drive them. For example, developers can deploy new GenAI tools to help write code documentation, increase productivity and modernise or convert classic mainframe code to languages such as Java and C#.
IT partners help better navigate the mainframe’s complexity
Despite the complexity of mainframe environments, more than half of Australian organisations expect their mainframe usage to increase in the next 12 months. Many have an eye toward the plethora of AI-powered use cases that can be used to transform them, so it makes sense that more than three quarters (77%) are using external providers to deliver mainframe modernisation projects.
The benefits of AI-enabled mainframe modernisation flow well beyond the IT department and support improvements across an entire organisation. For example, running AI models on the mainframe can provide the insights needed to enhance customer satisfaction — one of many reasons why AI is the secret ingredient that is transforming mainframes into intelligent powerhouses at the core of today’s enterprise.
