Compuware is rising from the ashes of its spinoffs. - Perhaps there really is money in mainframes

Compuware is a long-established and respected company specialising in supporting mainframes – which acquired and developed lots of interesting technology over time. Much of that technology was spun off in 2014. The remaining two business units were purchased by private equity firm Thoma Bravo a year or so back and then split into two distinct companies.

We all thought – including, I suspected Thoma Bravo – that when Compuware had spun out all its interesting technology, the mainframe business would stay around as a cash cow, gently declining as it exploited its locked-in and profitable mainframe customers.

How wrong we all were, I now think, and I shouldn’t be too surprised because the modern z Systems Mainframes are simply about the best designed and most innovative computing technology one can buy. People aren’t locked into the Mainframe by trickery, but by the fact that other technologies can’t really cope as well with the very biggest and most demanding applications. Nevertheless, although the latest z Mainframes may well be impressive and bang up-to-date, many Mainframe working practices are still back in the last century. Given how much money passes through Mainframe platforms, surely there must be a profit opportunity here?

Sam Knutson of Compuware has been explaining to me that, although Thoma Bravo’s immediate expectation of Compuware was guarded, Compuware’s new CEO, Chris O’Malley, was the catalyst for change, probably largely as a result of the time he’s spent in startups. He has now taken Compuware from a company with a “waterfall” focus, no new products in years, and updates every year or so, to a new Agile culture with quarterly deliveries. He delivered Compuware’s first new Mainframe product in ages.

Knutson emphasises that Linux on z is good to have – a “useful sideshow” – but that the real “mainframe edge” comes with zOS: CICS, IMS, MQ etc. That’s where the “gold nugget” is, he says, and points out that the z Mainframe is about the most advanced processing platform one can buy and that IBM has super compilers for this platform, plus tooling and dev process, even DevOps. Compuware is promising to revolutionise the Mainframe experience and innovate in the areas of process, tooling and DevOps.

The problem is, he says, that: “We have a top-class platform with last-century business processes and user mindset”.

He promises that Compuware really has started to address this. For example, it was clear to Compuware that Endevor, the widely-adopted CA Technologies legacy mainframe SCM tool, wasn’t a good enough fit for an Agile world, so Compuware tried out ISPW and liked it so much it bought it. And it has added static and dynamic analysis for code quality.

This is also going to be a cultural revolution “mainframe has changed but people still work the same way” – so Compuware now has “Town Hall” style meetings every two weeks, to help it manage its own cultural change.

The first new Compuware mainframe product for years is Topaz. This enables increasingly dynamic visualisation of Mainframe assets – “opening up the black box”, Compuware says – with, for example, the Topaz Runtime Visualiser. Compuware now sees visualisation as a key agent for productive disruption of any cosy Mainframe backwaters. Mainframe IDE Topaz Workbench, integrated with ISPW, supports multithreaded collaborative independent team working, which is centrally co-ordinated – not single-streamed working as in the 1980s.

Compuware has also learned something about changing conditions in the world of technology. Its VIP programme is all about applying modern analytics (an excellent idea) to, for example, usage information available on the Mainframe, leading to the discovery of effectiveness metrics and driving continuous improvement. This is about getting more out of what you already own, making Compuware into an effective technology partner instead of just a technology vendor. This approach includes determining internal maturity, compared to industry baselines, monitoring the effectiveness of developer tools in general and analysing customer churn.

It is also now delivering technology integrations with popular tools for web/mobile development so customers can use their tools of choice, rather than being locked into a monolithic toolset from a single vendor. For example, using SonarSource’s SonarQube for code quality management; we like the idea of using something established rather than reinventing the wheel.

Compuware has adopted a new focus on providing “customer experience”, which we welcome. It all sounds very good, but the proof of the pudding is in the eating and we look forward to new customer stories. Nevertheless, the signs are good and Knutson tells me that it hopes to have an effective and growing User Community in place by the end of 2016 – which, in part, would help to validate achievement of its new direction.