43 citations · 150 across the 13 of their papers we have counts for
13 papers
SigmaQuant: Hardware-Aware Heterogeneous Quantization Method for Edge DNN Inference
Qunyou Liu, Pengbo Yu, Marina Zapater +1
Deep neural networks (DNNs) are essential for performing advanced tasks on edge or mobile devices, yet their deployment is often hindered by severe resource constraints, including…
GreenLLM: SLO-Aware Dynamic Frequency Scaling for Energy-Efficient LLM Serving
Qunyou Liu, Darong Huang, Marina Zapater +1
Large Language Models (LLMs) are becoming the backbone of modern cloud services, yet their inference costs are dominated by GPU energy. Unlike traditional GPU workloads, LLM infere…
MatrixFlow: System-Accelerator co-design for high-performance transformer applications
Qunyou Liu, Marina Zapater, David Atienza
Transformers are central to advances in artificial intelligence (AI), excelling in fields ranging from computer vision to natural language processing. Despite their success, their…
Gem5-AcceSys: Enabling System-Level Exploration of Standard Interconnects for Novel Accelerators
Qunyou Liu, Marina Zapater, David Atienza
The growing demand for efficient, high-performance processing in machine learning (ML) and image processing has made hardware accelerators, such as GPUs and Data Streaming Accelera…
Enhancing Regression Models for Complex Systems Using Evolutionary Techniques for Feature Engineering
Patricia Arroba, José L. Risco-Martín, Marina Zapater +2
This work proposes an automatic methodology for modeling complex systems. Our methodology is based on the combination of Grammatical Evolution and classical regression to obtain an…
Reconsidering the performance of DEVS modeling and simulation environments using the DEVStone benchmark
José L. Risco-Martín, Saurabh Mittal, Juan Carlos Fabero +2
The Discrete Event System Specification formalism (DEVS), which supports hierarchical and modular model composition, has been widely used to understand, analyze and develop a varie…
LionHeart: A Layer-based Mapping Framework for Heterogeneous Systems with Analog In-Memory Computing Tiles
Corey Lammie, Yuxuan Wang, Flavio Ponzina +7
When arranged in a crossbar configuration, resistive memory devices can be used to execute Matrix-Vector Multiplications (MVMs), the most dominant operation of many Machine Learnin…
CloudProphet: A Machine Learning-Based Performance Prediction for Public Clouds
Darong Huang, Luis Costero, Ali Pahlevan +2
Computing servers have played a key role in developing and processing emerging compute-intensive applications in recent years. Consolidating multiple virtual machines (VMs) inside…
Runtime data center temperature prediction using Grammatical Evolution techniques
Marina Zapater, José L. Risco-Martín, Patricia Arroba +3
Data Centers are huge power consumers, both because of the energy required for computation and the cooling needed to keep servers below thermal redlining. The most common technique…
ALPINE: Analog In-Memory Acceleration with Tight Processor Integration for Deep Learning
Joshua Klein, Irem Boybat, Yasir Qureshi +6
Analog in-memory computing (AIMC) cores offers significant performance and energy benefits for neural network inference with respect to digital logic (e.g., CPUs). AIMCs accelerate…
The RECIPE Approach to Challenges in Deeply Heterogeneous High Performance Systems
Giovanni Agosta, William Fornaciari, David Atienza +8
RECIPE (REliable power and time-ConstraInts-aware Predictive management of heterogeneous Exascale systems) is a recently started project funded within the H2020 FETHPC programme, w…
Interpreting Deep Learning Models for Epileptic Seizure Detection on EEG signals
Valentin Gabeff, Tomas Teijeiro, Marina Zapater +4
While Deep Learning (DL) is often considered the state-of-the art for Artificial Intelligence-based medical decision support, it remains sparsely implemented in clinical practice a…
Graphene-based Wireless Agile Interconnects for Massive Heterogeneous Multi-chip Processors
Sergi Abadal, Robert Guirado, Hamidreza Taghvaee +18
The main design principles in computer architecture have recently shifted from a monolithic scaling-driven approach to the development of heterogeneous architectures that tightly c…