Machine learning migration for efficient near-data processing
AS Cordeiro, SR dos Santos, FB Moreira… - 2021 29th Euromicro …, 2021 - ieeexplore.ieee.org
Machine Learning (ML) rises as a highly useful tool to analyze the vast amount of data
generated in every field of science nowadays. Simultaneously, data movement inside …
generated in every field of science nowadays. Simultaneously, data movement inside …
Survey on near-data processing: Applications and architectures
One of the main challenges for modern processors is the data transfer between processor
and memory. Such data movement implies high latency and high energy consumption. In …
and memory. Such data movement implies high latency and high energy consumption. In …
Vector in memory architecture for simple and high efficiency computing
Data movement is one of the main challenges of contemporary system architectures. Near-
Data Processing (NDP) mitigates this issue by moving computation closer to the memory …
Data Processing (NDP) mitigates this issue by moving computation closer to the memory …
Efficient Machine Learning execution with Near-Data Processing
AS Cordeiro, SR dos Santos, FB Moreira… - Microprocessors and …, 2022 - Elsevier
A myriad of Machine Learning (ML) algorithms has emerged as a basis for many
applications due to the facility in obtaining satisfactory solutions to a wide range of problems …
applications due to the facility in obtaining satisfactory solutions to a wide range of problems …
On the SPEC-CPU 2017 opportunities for dynamic vectorization possibilities on PIM architectures
RM Sokulski, SR dos Santos… - Anais do XXIII Simpósio em …, 2022 - sol.sbc.org.br
Resumo Processing-In-Memory (PIM) devices usually implement vector instructions to
efficiently utilize the large main memory bandwidth. One possible way to vectorize …
efficiently utilize the large main memory bandwidth. One possible way to vectorize …
Advancing Database System Operators with Near-Data Processing
SR dos Santos, FB Moreira, TR Kepe… - 2022 30th Euromicro …, 2022 - ieeexplore.ieee.org
As applications become more data-intensive, issues like von Neumann's bottleneck and the
memory wall became more apparent since data movement is the main source of inefficiency …
memory wall became more apparent since data movement is the main source of inefficiency …
A computation-in-memory accelerator based on resistive devices
HAD Nguyen, J Yu, MA Lebdeh, M Taouil… - Proceedings of the …, 2019 - dl.acm.org
Today's computing architectures suffer from the three well-known bottlenecks, which are the
memory, the power and the instruction-level parallelism walls. Emerging non-volatile …
memory, the power and the instruction-level parallelism walls. Emerging non-volatile …
Video Decoder Improvements with Near-Data Speculative Motion Compensation Processing
G De Souza, JR Azambuja, B Zatt… - … on Circuits and …, 2022 - ieeexplore.ieee.org
Video decoder implementations are still evolving as they directly affect a large fraction of
embedded systems nowadays. In this context, Versatile Video Coding (VVC) brings …
embedded systems nowadays. In this context, Versatile Video Coding (VVC) brings …
Database Processing-in-Memory: A Vision
The recent trend of Processing-in-Memory (PIM) promises to tackle the memory and energy
wall problems lurking in the data movement around the memory hierarchy, like in data …
wall problems lurking in the data movement around the memory hierarchy, like in data …