acm-header
Sign In

Communications of the ACM

ACM TechNews

Darpa Wants Software That Adapts, Lasts Over 100 Years


View as: Print Mobile App Share:
Artists concept of software system components dynamically adapting to resource changes within an operational IT system.

A new program launched by the U.S. Defense Advanced Research Projects Agency aims to develop software systems that can endure more than 100 years.

Credit: U.S. Defense Advanced Research Projects Agency

The U.S. Defense Advanced Research Projects Agency (DARPA) recently announced the launch of the Building Resource Adaptive Software Systems (BRASS) project, which was launched to develop software systems that can adapt and survive for more than 100 years.

BRASS will develop the computational and algorithmic requirements for software systems and data to remain robust in excess of 100 years, which should significantly improve software resilience, reliability, and maintainability. The program will focus on the issues of high costs and frustration with current software systems that continue to grow in complexity.

BRASS wants new, clean-slate approaches resulting in the automated discovery of relationships between computations and the resources they utilize, as well as techniques to safely and dynamically incorporate optimized, tailored algorithms and implementations constructed in response to ecosystem changes.

"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 behavior of applications," says DARPA's Suresh Jagannathan.

The BRASS program will be divided into three 16-month phases focusing on reducing the time to repair vulnerabilities, allowing various syntactic and semantic forms of adaptation to be applied over large code bases, enabling adaptation to be generally applicable for a significant fraction of the code base, and reducing analytics and runtime monitoring overhead.

From Network World
View Full Article

Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account