Research

Current energy estimation tools lack the flexibility and precision required for accurate analysis across different layers of computing, including applications, machine learning architectures, and hardware architectures. CompJouleS addresses these limitations by combining top-down and bottom-up approaches to provide accurate and efficient energy estimates. The tool is modular, allowing for expansion to newer architectures, and incorporates a computational complexity calculator to analyze the workload of various operations for user-defined algorithms.

• CompJouleS enables energy analysis on five different type of architectures (CPU, GPU, FPGA, Hybrid & ASICs)
• It is a modular tool which allows users to expand the scope to newer architectures
• Analyzes energy cost of any user defined algorithm on various hardware architectures
• Tool will incorporate computational complexity calculator to analyze the workload of various operations for user defined algorithm
• Combines top-down and bottom-up approaches to estimate energy
• CompJouleS enables energy estimation of a networks quantized to lower precision