Challenges and Opportunities to Enable Large-Scale Computing via Heterogeneous Chiplets
Fast-evolving artificial intelligence (AI) algorithms such as large language models have
been driving the ever-increasing computing demands in today's data centers …
been driving the ever-increasing computing demands in today's data centers …
[PDF][PDF] Neuromorphic computing based on CMOS-integrated memristive arrays: current state and perspectives
AN Mikhaylov, EG Gryaznov… - Supercomputing …, 2023 - researchgate.net
The paper presents an analysis of current state and perspectives of high-performance
computing based on the principles of information storage and processing in biological …
computing based on the principles of information storage and processing in biological …
Spikesim: An end-to-end compute-in-memory hardware evaluation tool for benchmarking spiking neural networks
A Moitra, A Bhattacharjee, R Kuang… - … on Computer-Aided …, 2023 - ieeexplore.ieee.org
Spiking neural networks (SNNs) are an active research domain toward energy-efficient
machine intelligence. Compared to conventional artificial neural networks (ANNs), SNNs …
machine intelligence. Compared to conventional artificial neural networks (ANNs), SNNs …
REED: chiplet-based scalable hardware accelerator for fully homomorphic encryption
Fully Homomorphic Encryption (FHE) has emerged as a promising technology for
processing encrypted data without the need for decryption. Despite its potential, its practical …
processing encrypted data without the need for decryption. Despite its potential, its practical …
SWAP: A server-scale communication-aware chiplet-based manycore PIM accelerator
Processing-in-memory (PIM) is a promising technique to accelerate deep learning (DL)
workloads. Emerging DL workloads (eg, ResNet with 152 layers) consist of millions of …
workloads. Emerging DL workloads (eg, ResNet with 152 layers) consist of millions of …
Florets for Chiplets: Data Flow-aware High-Performance and Energy-efficient Network-on-Interposer for CNN Inference Tasks
H Sharma, L Pfromm, RO Topaloglu… - ACM Transactions on …, 2023 - dl.acm.org
Recent advances in 2.5 D chiplet platforms provide a new avenue for compact scale-out
implementations of emerging compute-and data-intensive applications including machine …
implementations of emerging compute-and data-intensive applications including machine …
COIN: Communication-aware in-memory acceleration for graph convolutional networks
Graph convolutional networks (GCNs) have shown remarkable learning capabilities when
processing graph-structured data found inherently in many application areas. GCNs …
processing graph-structured data found inherently in many application areas. GCNs …
Towards efficient in-memory computing hardware for quantized neural networks: state-of-the-art, open challenges and perspectives
O Krestinskaya, L Zhang… - IEEE Transactions on …, 2023 - ieeexplore.ieee.org
The amount of data processed in the cloud, the development of Internet-of-Things (IoT)
applications, and growing data privacy concerns force the transition from cloud-based to …
applications, and growing data privacy concerns force the transition from cloud-based to …
Architecture and application co-design for beyond-FPGA reconfigurable acceleration devices
In recent years, field-programmable gate arrays (FPGAs) have been increasingly deployed
in datacenters as programmable accelerators that can offer software-like flexibility and …
in datacenters as programmable accelerators that can offer software-like flexibility and …
A Heterogeneous Chiplet Architecture for Accelerating End-to-End Transformer Models
Transformers have revolutionized deep learning and generative modeling, enabling
unprecedented advancements in natural language processing tasks. However, the size of …
unprecedented advancements in natural language processing tasks. However, the size of …