A technical paper titled “A Precision-Optimized Fixed-Point Near-Memory Digital Processing Unit for Analog In-Memory Computing” was published by researchers at IBM Research Europe and IIS-ETH Zurich.
“Analog In-Memory Computing (AIMC) is an emerging technology for fast and energy-efficient Deep Learning (DL) inference. However, a certain amount of digital post-processing is required to deal with circuit mismatches and non-idealities associated with the memory devices. Efficient near-memory digital logic is critical to retain the high area/energy efficiency and low latency of AIMC. Existing systems adopt Floating Point 16 (FP16) arithmetic with limited parallelization capability and high latency. To overcome these limitations, we propose a Near-Memory digital Processing Unit (NMPU) based on fixed-point arithmetic. It achieves competitive accuracy and higher computing throughput than previous approaches while minimizing the area overhead. Moreover, the NMPU supports standard DL activation steps, such as ReLU and Batch Normalization. We perform a physical implementation of the NMPU design in a 14 nm CMOS technology and provide detailed performance, power, and area assessments. We validate the efficacy of the NMPU by using data from an AIMC chip and demonstrate that a simulated AIMC system with the proposed NMPU outperforms existing FP16-based implementations, providing 139× speed-up, 7.8× smaller area, and a competitive power consumption. Additionally, our approach achieves an inference accuracy of 86.65 %/65.06 %, with an accuracy drop of just 0.12 %/0.4 % compared to the FP16 baseline when benchmarked with ResNet9/ResNet32 networks trained on the CIFAR10/CIFAR100 datasets, respectively.”
Find the technical paper here. Published February 2024 (preprint).
Ferro, Elena, Athanasios Vasilopoulos, Corey Lammie, Manuel Le Gallo, Luca Benini, Irem Boybat, and Abu Sebastian. “A Precision-Optimized Fixed-Point Near-Memory Digital Processing Unit for Analog In-Memory Computing.” arXiv preprint arXiv:2402.07549 (2024).
Related Reading
3D Integration Supports CIM Versatility And Accuracy
Solving compute-in-memory’s limitations requires new approaches and dimensions.
Increasing AI Energy Efficiency With Compute In Memory
How to process zettascale workloads and stay within a fixed power budget.
Leave a Reply