HyperForest: A high performance multi-processor architecture for real-time intelligent systems
- Sandia National Labs., Albuquerque, NM (United States)
- Univ. of New Mexico, Albuquerque, NM (United States). Electrical Engineering and Computer Engineering Dept.
Intelligent Systems are characterized by the intensive use of computer power. The computer revolution of the last few years is what has made possible the development of the first generation of Intelligent Systems. Software for second generation Intelligent Systems will be more complex and will require more powerful computing engines in order to meet real-time constraints imposed by new robots, sensors, and applications. A multiprocessor architecture was developed that merges the advantages of message-passing and shared-memory structures: expendability and real-time compliance. The HyperForest architecture will provide an expandable real-time computing platform for computationally intensive Intelligent Systems and open the doors for the application of these systems to more complex tasks in environmental restoration and cleanup projects, flexible manufacturing systems, and DOE`s own production and disassembly activities.
- Research Organization:
- Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
- Sponsoring Organization:
- USDOE, Washington, DC (United States)
- DOE Contract Number:
- AC04-94AL85000
- OSTI ID:
- 465886
- Report Number(s):
- SAND-97-0843; ON: DE97004912; TRN: 97:002744
- Resource Relation:
- Other Information: PBD: Apr 1997
- Country of Publication:
- United States
- Language:
- English
Similar Records
Are RISCs ready for real-time control?
A model for an intelligent operating system for executing image understanding tasks on a reconfigurable parallel architecture