Monoidics applies mathematical analysis to software verification

Just as Stanford University has been critical to the success of Silicon Valley, Tech City’s viability as a technology start-up hub will depend in large part on its ability to access and commercialise the technology research conducted in London’s universities.

One example of deep computer science being put into practice in the Old Street area is Monoidics, a software verification provider whose technology is based on research at Queen Mary, University of London and Imperial College London.

According to founder Dr Dino Distefano, the technology can analyse complex software programs and assess whether or not they are likely to crash. It is based, he says, on mathematical analysis of the ‘pointers’ within a program – data fields that refer to other data points stored in memory.

Malfunctioning pointers – when a line of code refers to data stored in memory that does not exist – is the cause of 90% of program crashes, Distefano claims.

In a complex program, he says, it is very difficult for developers to keep track of whether all the pointers will function properly. “There are so many pointers that reference from one to another that the picture gets completely messy. Nobody knows what’s what.”

Put simply, Monoidics’s software analyses all the pointers in a program to ensure that they all resolve correctly.

The technology has been many years in gestation. “When we started, we could only analyse very small programs that you might find in a textbook. A few years later, we got it to work with real programs. At a certain point, we found the key to making the technique scale to large programs with millions of lines of code.

“When we got to that point, we thought: maybe we should commercialise this.”

Distefano and colleagues launched Monoidics, named after an esoteric algebraic structure, in 2009. It won its first contract through academic connections, and has grown from there without any outside investment. Customers include chip designer ARM Holdings, aircraft manufacturer Airbus and electrical engineering giant Mitsubishi Electric.

For an academic like Distefano, the move into business has taken some getting used to.

“For me, it’s a completely different world,” he says. “Getting a contract with one of these companies is a big pain because it takes a lot of interaction.”

He says setting up in the Old Street area allowed him to learn a lot from fellow start- ups in the region. “I know nothing about business, so I learned a lot by going to meetings for start-ups and so on.”

However, Distefano reports that there was not a lot of common ground between Monoidics and its more consumer-facing neighbours. “We are very techy compared to all the other start-ups,” he says. “It’s difficult to make an interaction and find common points with other people.”

Distefano says that the company is currently negotiating two major contracts, and if either one of them comes off it will have to double its capacity.

Monoidics has now appointed its own managing director, and Distefano is spending more time on his research than the company.

Still, he has plenty of ambition for the firm. “From my point of view as a scientist, I would like every engineer and every developer to use this technology, because I think it’s good.

“Wearing my businessman’s hat, I would say that I hope the company grows, and maybe at some point we can become part of a bigger company that can help us. “Either we grow big ourselves or we might be acquired by someone else.”

Related Topics