X
Innovation

New DARPA project aims to do away with IT updates

DARPA is hoping to do away with software updates as it embarks on a new project to establish a computer system designed to outlive 100 years of technological change.
Written by Leon Spencer, Contributor

The United States Defense Advanced Research Projects Agency (DARPA) is launching a program aimed at helping do away with software updates, with plans to design a computer system that has the ability to outlive over 100 years of technological change.

The four-year project, Building Resources Adaptive Software Systems (BRASS), for which DARPA is currently soliciting research proposals, will look at the computational and algorithmic requirements needed for software systems and accompanying data to remain functional for a century or longer.

According to the US Department of Defense technology research agency, the program is aimed at making advances in the design and implementation of long-lived software systems that have the ability to dynamically adapt to changes in the resources they depend upon, and the environments in which they operate.

DARPA, which played a vital role in establishing the protocol standards that led to the development of the internet, claims that such advances are likely to require new "linguistic abstractions" and resource-aware programs that are able to discover and specify program transformations, as well as systems designed to monitor changes in the surrounding digital ecosystem.

It is hoped that the program will lead to significant improvements in software resilience, reliability, and maintainability, ultimately resulting in the development of computer systems that can adapt to evolving operating conditions without the need for external update cycles.

"Technology inevitably evolves, but, very often, corresponding changes in libraries, data formats, protocols, input characteristics, and models of components in a software ecosystem undermine the behaviour of applications," said DARPA program manager Suresh Jagannathan. "The inability to seamlessly adapt to new operating conditions undermines productivity, hampers the development of cybersecure infrastructure, and raises the long-term risk that access to important digital content will be lost as the software that generates and interprets content becomes outdated."

DARPA claims that the premise on which the BRASS project will operate will be an entirely new "clean-slate" approach to software design, composition, and adaptation, and is hoped to result in algorithmic transformations that enable applications to adapt to changes without the need for extensive programmer involvement.

Jagannathan said that although the project's plan of ensuring applications continue to function correctly and efficiently in the face of a changing operational environment is a "formidable challenge", it is a challenge worth engaging in.

"Failure to respond to these changes can result in technically inferior and potentially vulnerable systems," said Jagannathan. "Equally concerning, the lack of automated upgrade mechanisms to restructure and transform applications leads to high software maintenance costs and premature obsolescence of otherwise functionally sound software."

While a number of computer hardware and software systems designed and constructed over 50 years ago are still in use today, many of these have required extensive updates and overhauls to remain relevant to external technological developments.

DARPA's new project comes as the COBOL programming language enters its 56th year of existence, with the language still being widely used in legacy applications by enterprises within the business and finance sectors, albeit after a handful of major revisions.

As recently as 2013, Micro Focus CTO Stuart McGill suggested that it was almost impossible for most people, in their day-to-day lives, to avoid a COBOL application.

"Normally, most transactions that we go through every day would be supported by COBOL applications, still are, have been for 30 to 40 years, probably still will be for 10 to 20 at least," McGill told ZDNet in November 2013.

Editorial standards