U.S. scientists say they have created an automated program designed to "debug" the nation's nuclear test computer simulations.

Purdue University researchers, working with scientists at the Lawrence Livermore National Laboratory, said their program, called AutomaDeD (pronounced like automated), finds errors in computer code for complex "parallel" programs.

Because international treaties forbid the detonation of nuclear test weapons, certification is done using complex simulations. The researchers said such simulations, which may contain as many as 100,000 lines of computer code, must accurately show reactions taking place on the scale of milliseconds, or thousandths of a second.

"The simulations take several weeks to run, and then they have to be debugged to correct errors in the code," said Associate Professor Saurabh Bagchi. "The error might have occurred in the first hour of operation, and if you had known about it you could have stopped it then. The idea is to use AutomaDeD as the simulation is running to automatically monitor what's happening.

"If things start going wrong, AutomaDeD would stop and flag which process and which part of the code in the process is likely anomalous."

Recent research findings show AutomatDeD was 90 percent accurate in identifying the time "phase" when bugs occurred, 80 percent accurate in identifying the specific tasks involved and 70 percent accurate in identifying the interference faults.

The research will be presented June 30 in Chicago during the 40th Annual IEEE/IFIP International Conference on Dependable Systems and Networks.