5 papers
NIFA: Nonlinear IMC enhanced FPGA for efficient ML inference
Jiajun Hu, Ruthwik Reddy Sunketa, Lei Zhao +3
Recent FPGAs have improved deep learning (DL) inference efficiency through dedicated tensor blocks and in-BRAM computation. ReRAM-based analog in-memory computing (IMC) pushes effi…
Boosting FPGA Performance with Direct BRAM-DSP Paths
Jiajun Hu, Ruthwik Reddy Sunketa, Andrew Boutros +1
Efficient data movement between memory and compute units is a key performance bottleneck in modern FPGA designs, particularly for deep learning (DL) workloads. In typical FPGA arch…
CarbonPATH: Carbon-aware pathfinding and architecture optimization for chiplet-based AI systems
Chetan Choppali Sudarshan, Jiajun Hu, Aman Arora +1
The exponential growth of AI has created unprecedented demand for computational resources, pushing chip designs to the limit while simultaneously escalating the environmental footp…
RACAM: Enhancing DRAM with Reuse-Aware Computation and Automated Mapping for ML Inference
Siyuan Ma, Jiajun Hu, Jeeho Ryoo +2
In-DRAM Processing-In-Memory (DRAM-PIM) has emerged as a promising approach to accelerate memory-intensive workloads by mitigating data transfer overhead between DRAM and the host…
CarbonSet: A Dataset to Analyze Trends and Benchmark the Sustainability of CPUs and GPUs
Jiajun Hu, Chetan Choppali Sudarshan, Vidya A. Chhabria +1
Over the years, the chip industry has consistently developed high-performance processors to address the increasing demands across diverse applications. However, the rapid expansion…