It’s not only the way applications are used that is outdated, but the industry needs to make fundamental changes in the way they are developed as well.
That’s the word from SAP’s member of the executive board: products and innovation Dr Vishal Sikka, who describes application development as primitive and ripe for change.
Speaking at the opening of SAP’s new Innovation Centre in Potsdam, Germany, Dr Sikka says spaces are important to the innovative process, hence the company’s development of app places, innovation centres and now HANA cafes, all designed to fire the creative spirit.
“The words structured and innovative don’t go together,” he says. “Innovation has to happen in the open, in a free style.”
Although new ideas and projects in general will be handled at the new innovation centre, Dr Sikka says the facility has teams focusing on two main areas: applications development and healthcare.
“The notion of the application as a game changer has taken off,” he says. “But the way we build these applications has to change as well.”
Dr Sikka goes so far as to describe the current state of application development as being in crisis. For instance, he cites as an example the fact that many financial institutions are basing their businesses on core software that is at least a decade old. In addition, these systems are disconnected and battle to share data with one another. This means that banks have a real problem in assessing risks based on reliable data.
Retail companies, because of a similar problem of old and disconnected systems, cannot get a clear picture of their customers.
Even more recent systems, built by oil and gas companies, cannot perform fast enough to process up to date meteorological data.
“Just think, though, if we had better software development tools, we could develop software for any of these problems.”
Building on the foundation of HANA, Dr Sikka believes these and many other problems could be addressed. With a rethink in the way we develop software, performance could be dramatically improved and the breakdown in the way we manage applications could be fixed as well.
As to the nuts and bolts of how to achieve this, Dr Sikka points out that software has spent some years trying to get rid of batch processes – but software development itself is a batch process.
“Teams of individual developers write code, then check it into a central repository. There it is checked, compiled and debugged – usually by a different team, so there is no communication during this process. We can’t even simulate the code using real data.
“What if you could see all the lines of code as you write your own; it you could talk to the other people working on the same application, writing code in parallel with them?”
A very real problem with much of the software in use today is that the applications tend to have outlived their programming languages, Dr Sikka points out. This means companies still rely on systems written in languages like Cobol, which are difficult to maintain largely because of a shortage of relevant skills.
“In-memory computing makes this an easy problem to solve,” he says. “It allows us to articulate an application in a way that is independent of the programming language and allows us to bridge the divide between abstraction and performance.”
The other great opportunity – and challenge – facing software development today lies in solving the great problems in the world, Dr Sikka says. These revolve around many issues including education, healthcare and the monetary system.
Looking ahead to new hardware breakthroughs, Dr Sikka believes silicon photonic interconnects will soon allow hardware to scale for higher performance, while non-volatile memory will help to make in-memory computing the default standard for databases.
Apart from Potsdam, SAP innovation centres either already exist or are planned for Berlin, New York, Shanghai, Beijing, Bangalore, Delhi and Paris.