publications

2024

1. DATE 2024

   NOVA: NoC-based Vector Unit for Mapping Attention Layers on a CNN Accelerator

   Upadhyay, Mohit, Juneja, Rohan, Wang, Bo, Zhou, Jun, Wong, Weng-Fai, and Peh, Li-Shiuan

   In Proceedings of the 2024 Conference & Exhibition on Design, Automation & Test in Europe 2024

   Abs

   Attention mechanisms are becoming increasingly popular, being used in neural network models in multiple domains such as natural language processing (NLP) and vision applications, especially at the edge. However, attention layers are difficult to map onto existing neuro accelerators since they have a much higher density of non-linear operations, which lead to inefficient utilization of today’s vector units. This work introduces NOVA, a NoC-based Vector Unit that can perform non-linear operations within the NoC of the accelerators, and can be overlaid onto existing neuro accelerators to map attention layers at the edge. Our results show that the NOVA architecture is up to 37.8x more power-efficient than state-of-the-art hardware approximators when running existing attention-based neural networks.

2023

1. ISCAS 2023

   1.7pJ/SOP Neuromorphic Processor with Integrated Partial Sum Routers for In-Network Computing

   Wang, B., Wong, M. M., Li, D., Chong, Y. S., Zhou, J., Wong, W. F., Peh, L., Mani, A., Upadhyay, M., Balaji, A., and Do, A. T.

   In 2023 IEEE International Symposium on Circuits and Systems (ISCAS) 2023

2022

1. ISOCC 2022

   Network-on-Chip-Centric Accelerator Architectures for Edge AI Computing

   Wang, Bo, Dong, Ke, Zakaria, Nurul Akhira Binte,  Upadhyay, Mohit, Wong, Weng-Fai, and Peh, Li-Shiuan

   In 2022 19th International SoC Design Conference (ISOCC) 2022
2. ICDCS 2022

   Blockchain Machine: A Network-Attached Hardware Accelerator for Hyperledger Fabric

   Javaid, Haris, Yang, Ji, Santoso, Nathania,  Upadhyay, Mohit, Mohan, Sundararajarao, Hu, Chengchen, and Brebner, Gordon

   In 2022 IEEE 42nd International Conference on Distributed Computing Systems (ICDCS) 2022
3. DAC 2022

   REACT: A Heterogeneous Reconfigurable Neural Network Accelerator with Software-Configurable NoCs for Training and Inference on Wearables

   Upadhyay, Mohit, Juneja, Rohan, Wang, Bo, Zhou, Jun, Wong, Weng-Fai, and Peh, Li-Shiuan

   In Proceedings of the 59th ACM/IEEE Design Automation Conference 2022

   Abs

   On-chip training improves model accuracy on personalised user data and preserves privacy. This work proposes REACT, an AI accelerator for wearables that has heterogeneous cores supporting both training and inference. REACT’s architecture is NoC-centric, with weights, features and gradients distributed across cores, accessed and computed efficiently through software-configurable NoCs. Unlike conventional dynamic NoCs, REACT’s NoCs have no buffer queues, flow control or routing, as they are entirely configured by software for each neural network. REACT’s online learning realises upto 75% accuracy improvement, and is upto 25x faster and 520x more energy-efficient than state-of-the-art accelerators with similar memory and computation footprint.

2019

1. VLSID 2019

   Multi-application Based Network-on-Chip Design for Mesh-of-Tree Topology Using Global Mapping and Reconfigurable Architecture

   Upadhyay, Mohit, Shah, Monil, Bhanu, P. Veda, Soumya, J., and Cenkeramaddi, Linga Reddy

   In 2019 32nd International Conference on VLSI Design and 2019 18th International Conference on Embedded Systems (VLSID) 2019
2. VDAT 2018

   A Novel Fault-Tolerant Routing Algorithm for Mesh-of-Tree Based Network-on-Chips

   Shah, Monil,  Upadhyay, Mohit, Bhanu, P. Veda, Soumya, J., and Cenkeramaddi, Linga Reddy

   In VLSI Design and Test 2019

2018

1. HPCS 2018

   Fault Tolerant Routing Methodology for Mesh-of-Tree based Network-on-Chips using Local Reconfiguration

   Upadhyay, Mohit, Shah, Monil, Bhanu, P. Veda, Soumya, J., and Cenkeramaddi, Linga Reddy

   In 2018 International Conference on High Performance Computing Simulation (HPCS) 2018
2. TENCON 2018

   A Novel Fault-Tolerant Routing Technique for Mesh-of-Tree based Network-on-Chip Design

   Upadhyay, Mohit, Shah, Monil, Bhanu, P. Veda, J, Soumya, Cenkarmaddi, Linga Reddy, and Idsøe, Henning

   In TENCON 2018 - 2018 IEEE Region 10 Conference 2018