MIMDRAM: An End-to-End Processing-Using-DRAM System for High-Throughput, Energy-Efficient and Programmer-Transparent Multiple-Instruction Multiple-Data …
Processing-using-DRAM (PUD) is a processing-in-memory (PIM) approach that uses a
DRAM array's massive internal parallelism to execute very-wide (eg, 16,384-262,144-bit …
DRAM array's massive internal parallelism to execute very-wide (eg, 16,384-262,144-bit …
PIM-GPT: A Hybrid Process-in-Memory Accelerator for Autoregressive Transformers
Decoder-only Transformer models such as GPT have demonstrated superior performance in
text generation, by autoregressively predicting the next token. However, the performance of …
text generation, by autoregressively predicting the next token. However, the performance of …
Accelerating Graph Neural Networks on Real Processing-In-Memory Systems
C Giannoula, P Yang, IF Vega, J Yang, YX Li… - arXiv preprint arXiv …, 2024 - arxiv.org
Graph Neural Networks (GNNs) are emerging ML models to analyze graph-structure data.
Graph Neural Network (GNN) execution involves both compute-intensive and memory …
Graph Neural Network (GNN) execution involves both compute-intensive and memory …
Simplepim: A software framework for productive and efficient processing-in-memory
Data movement between memory and processors is a major bottleneck in modern
computing systems. The processing-in-memory (PIM) paradigm aims to alleviate this …
computing systems. The processing-in-memory (PIM) paradigm aims to alleviate this …
Evaluating Homomorphic Operations on a Real-World Processing-In-Memory System
H Gupta, M Kabra, J Gómez-Luna… - 2023 IEEE …, 2023 - ieeexplore.ieee.org
Computing on encrypted data is a promising approach to reduce data security and privacy
risks, with homomorphic encryption serving as a facilitator in achieving this goal. In this work …
risks, with homomorphic encryption serving as a facilitator in achieving this goal. In this work …
Machine learning training on a real processing-in-memory system
Machine learning (ML) algorithms [1]–[6] have become ubiquitous in many fields of science
and technology due to their ability to learn from and improve with experience with minimal …
and technology due to their ability to learn from and improve with experience with minimal …
Memory-centric computing
O Mutlu - arXiv preprint arXiv:2305.20000, 2023 - arxiv.org
Memory-centric computing aims to enable computation capability in and near all places
where data is generated and stored. As such, it can greatly reduce the large negative …
where data is generated and stored. As such, it can greatly reduce the large negative …
SwiftRL: Towards Efficient Reinforcement Learning on Real Processing-In-Memory Systems
Reinforcement Learning (RL) trains agents to learn optimal behavior by maximizing reward
signals from experience datasets. However, RL training often faces memory limitations …
signals from experience datasets. However, RL training often faces memory limitations …
Analysis of Distributed Optimization Algorithms on a Real Processing-In-Memory System
S Rhyner, H Luo, J Gómez-Luna, M Sadrosadati… - arXiv preprint arXiv …, 2024 - arxiv.org
Machine Learning (ML) training on large-scale datasets is a very expensive and time-
consuming workload. Processor-centric architectures (eg, CPU, GPU) commonly used for …
consuming workload. Processor-centric architectures (eg, CPU, GPU) commonly used for …
Scaling Down to Scale Up: A Cost-Benefit Analysis of Replacing OpenAI's GPT-4 with Self-Hosted Open Source SLMs in Production
C Irugalbandara, A Mahendra, R Daynauth… - arXiv preprint arXiv …, 2023 - arxiv.org
Many companies rely on APIs of managed AI models such as OpenAI's GPT-4 to create AI-
enabled experiences in their products. Along with the benefits of ease of use and shortened …
enabled experiences in their products. Along with the benefits of ease of use and shortened …