Researchers from the U.S. Army Research Laboratory’s Computational and Information Sciences Directorate (CISD) – David Richie and James Ross – won first place in an international software contest for their work on emulators. Their submission, Cycle-Accurate 8080 Emulation Using an ARM11 Processor with Dynamic Binary Translation, addresses some of the programming challenges of next generation computing devices.

The objective of the ACM-IEEE MEMOCODE software design contest is to make a Space Invaders video game emulator run faster on the Raspberry PI hardware than on its original platform (an Intel 8080 CPU) using only software solutions.

Unlocking the full potential of emerging heterogeneous computer systems is important for a range of applications, including the performance of mobile hand-held computing devices so crucial to the modern military.

The ARL researchers set out to enable cycle-accurate emulation of an Intel 8080 CPU using a low-power ARM11 CPU. First they experimented with software optimization techniques and threaded dispatch designs, then they applied dynamic binary translation between the two Instruction Set Architectures. With this technique, the stream of machine instructions from the original program is translated directly into instructions for the target processor when the program is run. The cycle efficiency of this design was nearly as fast as the original 8080 processor, according to Ross.

“The ability to emulate one computer with another efficiently allows the user-experience to be transparent,” said Richie. “The software developed for the contest translates machine instructions in a particularly fast and accurate manner. By improving the performance and accuracy of the translated instructions, the software can be run in a very energy-efficient manner.”

David Richie, President of Brown Deer Technology, and James Ross, a computational scientist with Engility Corporation, work closely with the other members of the CISD team, including Dale Shires, chief of ARL’s Computing Architectures Branch, Computational Sciences Division in CISD, and Song Park, CISD team leader.

As Song explains, “the project establishes foundation for multiple promising future work areas in software portability, dynamic optimization and binary parallelization that can support new architectures. In the long term, we envision enabling individual Soldiers with ubiquitous and intelligent computing.”

Ross and Richie will receive their award during the 12th ACM-IEEE International Conference on Formal Methods and Models for System Design (MEMOCODE’14), scheduled to be held in Switzerland, October 19-21, 2014.