5 papers
HeTraX: Energy Efficient 3D Heterogeneous Manycore Architecture for Transformer Acceleration
Pratyush Dhingra, Janardhan Rao Doppa, Partha Pratim Pande
Transformers have revolutionized deep learning and generative modeling to enable unprecedented advancements in natural language processing tasks and beyond. However, designing hard…
Dataflow-Aware PIM-Enabled Manycore Architecture for Deep Learning Workloads
Harsh Sharma, Gaurav Narang, Janardhan Rao Doppa +2
Processing-in-memory (PIM) has emerged as an enabler for the energy-efficient and high-performance acceleration of deep learning (DL) workloads. Resistive random-access memory (ReR…
FARe: Fault-Aware GNN Training on ReRAM-based PIM Accelerators
Pratyush Dhingra, Chukwufumnanya Ogbogu, Biresh Kumar Joardar +3
Resistive random-access memory (ReRAM)-based processing-in-memory (PIM) architecture is an attractive solution for training Graph Neural Networks (GNNs) on edge platforms. However,…
Block-Wise Mixed-Precision Quantization: Enabling High Efficiency for Practical ReRAM-based DNN Accelerators
Xueying Wu, Edward Hanson, Nansu Wang +9
Resistive random access memory (ReRAM)-based processing-in-memory (PIM) architectures have demonstrated great potential to accelerate Deep Neural Network (DNN) training/inference.…
Design-Space Exploration and Optimization of an Energy-Efficient and Reliable 3D Small-world Network-on-Chip
Sourav Das, Janardhan Rao Doppa, Partha Pratim Pande +1
A three-dimensional (3D) Network-on-Chip (NoC) enables the design of high performance and low power many-core chips. Existing 3D NoCs are inadequate for meeting the ever-increasing…