Lance arrived at LzLabs, the mainframe modernisation experts, two years ago on a mission to apply fresh thinking to tenacious problems that dog Fortune 500 companies. “I’m not a mainframe man”, he explains. Instead of relying on mainframe methods and skills to fix legacy problems, he orchestrates LzLabs’ diverse expertise to solve legacy integration problems in new ways. And, he reminds us, the ethos of ‘Laboratory’, contained in the LzLabs brand, continues to inform diverse efforts and solutions.
Original thinking is urgently needed as mainframes persist in IT estates, chosen once upon a time for their reliability and performance but presenting a significant barrier to modernisation today. A hefty 70% of blue-chip companies depend on mainframes for their critical business functions – collectively, they crunch 30 billion transactions every day, according to Thomson Data. However, the cost of maintaining mainframes is eye-watering, and data remains walled off from modern tools that reside in the cloud.
The key to the transformational magic that LzLabs dispenses in its migration programmes, which Borvansky channels in his Head of Development role, is variety. Incrementally moving mainframe assets intact onto modern platforms is the bread-and-butter work of the LzLabs Software Defined Mainframe® architecture. However, unique integration challenges occur along each client’s journey, and this is where diversity in the development team generates innovative thinking.
Thinking outside the legacy silo
Lance’s career spans a master’s in robotics at MIT, entrepreneurship, a stint in trading and finance, and, latterly, a deep dive into coding. Not being from the mainframe world can be an advantage, he says. “I’m not wedded to that worldview. Where I came from, things were changing all the time; software and the boxes were constantly swapped in and out”. A stark contrast to the mainframe universe that is powered by a computing constant and which is backwards compatible.
If anyone has lingering doubts about the value of embedding diversity within organisations, look no further than the Google team behind Generative AI. “A Welshman, a Ukrainian, an Indian woman, and a German meet at Google” may sound like the start of a bad joke. In fact, the combined linguistic brain power of a multi-national team of Google scientists devised a new way to process natural language based on relationships rather than sequences.
A chance encounter in a hallway at the search giant and a chat about the recent release of a science fiction film, “Arrival”, led to a discussion about how language could be analysed and generated differently. The extraterrestrials in the movie did not use linear language but units of speech or symbols, which humans had to decode. The notion of ‘self-attention’ was born—a Cambrian moment for AI, which also emphatically proved the value of diversity.
The right attitude, expertise, and track record build trust
Closer to home, a wide and deep skills pool at LzLabs conjured creative thinking at an automotive giant. The carmaker had a substantial portfolio of legacy COBOL, PL1, and assembler applications, which it decided to recode in Java when the mainframe skills shortage became acute. Two technical hurdles jeopardised the migrations: lack of data availability between refactored and legacy applications, and the refactoring of Assembler applications couldn’t be automated.
In response, the LzLabs team customised a relational database component and built a hybrid data access model. This, in turn, ensured data availability while minimising synchronisation risks. For the second problem, the LzLabs Software Defined Mainframe® (SDM) showed its adaptability in addressing interoperability needs between old and new languages. Selected Assembler modules were converted to Java via the SDM, which retained the ability to talk back to the remaining Assembler applications on the runtime environment.
It’s this kind of problem-solving that wins over customers and strengthens their partnership with LzLabs as they move along the mainframe modernisation journey together. Such trust, rooted in successful, incremental migratory steps, is essential in moving enterprises forward where there’s a barrier of fear and reluctance. According to researchers ISG, a massive 56 percent of 164 respondents surveyed in 2023 are either dissatisfied or have been burnt by their experience of mainframe migration.
“Companies have found workarounds to historic data integration issues,” Lance confirms. Once you start digging into people’s infrastructure, the spaghetti can be very tangled indeed, he adds. Multiple connections, systems, and databases have accumulated over the past decades. In enterprise IT, there have been numerous fashions of mainframes, Linux, then Salesforce —and there are a lot of residues. People ended up building on top.”
Bi-directional vision and teamwork power mainframe modernisation
It’s why Lance, who’s justifiably proud of the technical achievements of his team, talks in business vocabulary and frames the imperative for modernisation in the language of risk. “My claim is we help reduce complexity—and therefore risk—by getting off the mainframe. The mission is to get onto a common platform and enable access to data and information more easily. Organisations are then ready for the next stage. With a clear view of their data, they can access intelligence and tools and take the risks that make sense for their business”.
Gaining a clear view of the business is the quest of multi-visioned Lance and his team. Alongside the confidence built on having your problems solved by a highly talented, diverse, and experimental team is the trust that it banks. Building an ecosystem of trust that supports customers along all stages of the mainframe migration journey is an essential part of the LzLabs proposition.
