Why AI on IBM Z? Enhancing Mainframe Security, Efficiency, and Modernization
Bringing Mainframes Into the 21st Century
Once the rock stars of the IT world, mainframes have slowly drifted into the background. Yet, they continue to operate quietly behind the scenes, delivering on their original value: rock-solid stability, security, and resilience to support critical workloads for large enterprises. Today, mainframes handle more than 70% of the world’s transactional workloads for top airlines, retailers, banks, and government agencies. And 43 of the world’s top 50 banks and 8 of the top 10 payment companies continue to rely on mainframes as their core computing platform.
According to recent research by the IBM Institute for Business Value, 75% of global IT executives say that “mainframes are equal to or better than cloud computing in terms of total cost of ownership, indicating they remain a cost-effective option for optimizing IT budgets.” For these organizations, mainframes are strategic assets rather than technical burdens, which explains why mainframe owners continue to invest in them. Moving forward, IT leaders would like their mainframes to support continuous improvement and continuous delivery. And as COBOL expertise ages out of the workplace, IT leaders are looking for the best ways to refactor COBOL applications into modern programming languages.
In addition, as enterprises push ahead with digital and business transformation initiatives, IT leaders are finding that mainframes have a place at the table as an effective way to infuse AI into workloads and host enterprise data to support hybrid cloud architectures. Seventy-nine percent of IT executives say that mainframes are essential for unlocking the full potential of AI-driven business transformation. IT executives are now seeing their mainframes as a solid foundation for pairing AI with transactional workloads and mainframe operations to accelerate modernization and development.
With IBM’s launch of its Telum processor for IBM Z, IT leaders have a much more compelling reason to embrace AI for transaction processing and other crucial workloads where data resides on the mainframe and where security and reliability protect the backbone of the business. The fast Telum processor enables low-latency AI inference, making IBM Z ideal for providing real-time transaction processing and real-time decision-making. And by putting AI inference on IBM Z, organizations can infuse AI into core business workloads without buying additional infrastructure while getting the security privacy benefits of the mainframe.
Common Use Cases for AI on IBM Z
Modernize Legacy Code to Fill the Skills Gap
AI-enhanced application modernization tools are helping organizations close mainframe skills gaps by enabling developers to modernize legacy applications much faster and more efficiently than current methods.
Mainframe applications are often written in COBOL or other legacy programming languages, and can be difficult to decipher, especially when the code isn’t well commented. GenAI powered by watsonx Code Assistant for Z can quickly analyze code structure, build control program flow charts, and add comments to help developers who are more familiar with other languages better understand an application and its dependencies, which dramatically accelerates the code modernization process and helps developers quickly create high-quality, easy-to-maintain code.
Watsonx Code Assistant for Z uses the IBM Granite large language model (LLM), which is purpose-built for understanding, explaining, and refactoring legacy code like COBOL into a modern language such as Java. Developers who attempt to do this with a public LLM such as ChatGPT will end with “JOBOL,” which doesn’t fully leverage the benefits of modern Java, limiting the success of your modernization efforts. Enterprises need more precision to make code modernization work for them.
Analyze for Fraud in Real Time
Financial organizations have been using AI for years to analyze transactions to detect fraud. While these machine learning algorithms have been useful, enforcement has been piecemeal because fraud detection has been limited to analyzing random samples.
With AI on IBM Z and the Telum processor, latency is so low that financial companies can now detect fraud in real time and analyze 100% of transactions with fewer false positives, leading to better customer satisfaction and fewer payouts, for example, for fraudulent credit card transactions.
A Foundation for Digital Transformation and Hybrid Cloud
Mainframes are also becoming solid foundations for hybrid cloud environments to simplify the growing complexity of integrating data and applications and handle the explosive growth of data across multiple systems. Mainframes can accelerate digital and business transformation by integrating on-premise data with public cloud services. For example, many enterprises are adopting a data-centric strategy that syncs mainframe application data with the cloud. By improving how mainframes integrate with other computing assets, organizations can more fully realize the potential of their enterprise data to support AI-driven use cases.
Simplify Mainframe Operations
AI on IBM Z can also improve mainframe management to help busy operations professionals keep up with the demands of modernization. AI-powered automation, predictive analytics, self-healing, and self-tuning capabilities can proactively detect and prevent issues, optimize workflows, and improve system reliability. As mainframe operations expertise retires, IBM can help organizations fill the skills gap with AI.
Security Advantages of AI on IBM Z
Your Data Stays Private
While tools like ChatGPT have opened new doors for companies, the use of public GenAI to support business operations has remained off-limits due in part to the risk of exposing sensitive information to a public platform.
A key advantage of AI on IBM Z is that the IBM Granite LLMs don’t use your data for training or any other purpose, so the risk of exposing sensitive data is nearly zero. When organizations feed legacy code to an IBM LLM, IBM retains nothing. This is part of IBM’s commitment to its customers and is another important reason why AI on IBM Z is attractive to customers.
Resilient to Attack
Enhanced data security is another important reason to count on mainframes to store and process sensitive data. While the average cost of a data breach continues to escalate, only 0.1% of mainframe customers have experienced a successful cyberattack. Mainframes can also use AI to identify shadow data, monitor for data access abnormalities, and alert cybersecurity teams about threats, with the ability to detect and remediate issues in real time.
In addition, hackers are intercepting and storing vast amounts of encrypted data with the intent to decrypt it in the future when quantum computing becomes viable. Modern mainframes encrypt data with quantum-safe algorithms to keep vital information safe for the future.
A Rock-solid Foundation for the Future
Unlike most other aging rock stars, IBM Z has been improved over time to remain relevant and vital. As organizations continue to look for ways to secure their data and innovate with AI, mainframes will continue to provide the rock-solid security and performance needed to make AI practical in the real world.
How Evolving Solutions Can Help
Getting a mainframe ready for AI takes time and planning. Clients can avoid common pitfalls and save time by partnering with Evolving Solutions to help implement AI for IBM Z at your organization. We’ve done this already for others and we have the experience to take you through a smooth journey to getting the business benefits of AI. We’ll help you get current with operating system and middleware software versions, recommend database changes to support an AI-driven environment, and deliver an implementation plan around your high-priority use cases for AI on IBM Z.
Contact Evolving Solutions today to get started.
