Showing papers by "Richard Lethin published in 2008"
01 Jan 2008
TL;DR: The Tera-op Reliable Intelligently Adaptive Processing System (TRIPS) as discussed by the authors is a full-system TRIPS prototype including a new Explicit Data Graph Execution (EDGE) instruction set architecture, custom application-specific integrated circuit (ASIC) chips, system circuit boards, a custom compiler with new optimization capabilities, a software development kit, and support for multithreaded parallel programs.
Abstract: Abstract : The Tera-op Reliable Intelligently Adaptive Processing System (TRIPS) is a novel computer system designed to address technology scaling challenges, to provide high performance through concurrency, and to demonstrate mechanisms for hardware polymorphism. The team has constructed a full-system TRIPS prototype including a new Explicit Data Graph Execution (EDGE) instruction set architecture, custom application-specific integrated circuit (ASIC) chips, system circuit boards, a custom compiler with new optimization capabilities, a software development kit, and support for multithreaded parallel programs. Consisting of approximately 170 million transistors in a 130nm technology, the TRIPS chip includes two coarse grained processors, each with 16 ALUs (including floating-point units) that execute in parallel. TRIPS systems of up to 20 chips have been deployed at UT-Austin, ISI-East, and the Air Force Research Laboratory (AFRL). The prototype demonstrates per-processor performance (measured in cycles) of up to three times better than leading commercial products. Performance analysis results have led to follow-on architectures that employ dynamic polymorphism to tailor the capabilities of the hardware to the demands of the software.
2 citations
01 May 2008
TL;DR: The team investigated essential "Cognitive Information Processing" aspects of the architecture ranging from pure cognitive research into the essentials of human level reasoning to computing infrastructures that are essential to be able to transition PCAA into application domains that have a crucial need for this technology.
Abstract: : Lockheed Martin Advanced Technology Laboratories (LM-ATL) led a highly qualified and motivated team to research, develop, and validate the Polymorphous Cognitive Agent Architecture (PCAA) as part of DARPA's Architecture for Cognitive Information Processing (ACIP). Our team investigated essential "Cognitive Information Processing" aspects of the architecture ranging from pure cognitive research into the essentials of human level reasoning to computing infrastructures that are essential to be able to transition PCAA into application domains that have a crucial need for this technology. PCAA is a dynamic, adaptive cognitive architecture that makes previously intractable approximation tasks tractable for NP-hard cognitive problems. PCAA consists of: linear composable cognitive agents, a cognitive mark-up language for cognitive behavior definition, a cognitive layer for derivation of cognitive services and specialized cognitive agents, and a next generation polymorphic hardware and software layer for runtime composition and instantiation of cognitive agents.
1 citations