Dark Silicon Accelerators for Database Indexing

À propos
Confidentialité
Mentions légales

Graph Chatbot

Publications associées (27)

Low-Power Artificial Neural Network Perceptron Based on Monolayer MoS2

Andras Kis, Aleksandra Radenovic, Zhenyu Wang, Guilherme Migliato Marega, Mukesh Kumar Tripathi

Machine learning and signal processing on the edge are poised to influence our everyday lives with devices that will learn and infer from data generated by smart sensors and other devices for the Internet of Things. The next leap toward ubiquitous electron ...

2022

New Technologies to Enhance the Figures-of-Merit of GaN Power Devices

Luca Nela

Gallium Nitride (GaN) has enabled groundbreaking developments in the field of optoelectronics and radio frequency communication. More recently, GaN devices for power conversion applications have demonstrated excellent potential. Thanks to Gallium Nitride w ...

EPFL2022

Micro-architectural Analysis of Database Workloads

Utku Sirin

Database workloads have significantly evolved in the past twenty years. Traditional database systems that are mainly used to serve Online Transactional Processing (OLTP) workloads evolved into specialized database systems that are optimized for particular ...

EPFL2021

Relative stability toward diffeomorphisms indicates performance in deep nets

Mario Geiger, Matthieu Wyart, Leonardo Petrini, Alessandro Favero

Understanding why deep nets can classify data in large dimensions remains a challenge. It has been proposed that they do so by becoming stable to diffeomorphisms, yet existing empirical measurements support that it is often not the case. We revisit this qu ...

2021

Equinox: Training (for Free) on a Custom Inference Accelerator

Babak Falsafi, Ahmet Caner Yüzügüler, Mario Paulo Drumond Lages De Oliveira, Arash Pourhabibi Zarandi, Louis Coulon, Martin Jaggi

DNN inference accelerators executing online services exhibit low average loads because of service demand variability, leading to poor resource utilization. Unfortunately, reclaiming idle inference cycles is difficult as other workloads can not execute on a ...

ACM2021

Pushing the Limits of Efficiency and Power Density in High-Frequency Power Conversion Based on Wide-Band-Gap Technologies

Armin Jafari

The emergence of wide-band-gap (WBG) power transistors with low conduction losses and high-speed switching speeds has paved the way for more-than-ever efficient power electronics systems and huge energy saving potentials. Likewise, power density- the ratio ...

EPFL2021

Resonances on GaN-on-Si Epitaxies: A Source of Output Capacitance Losses in Power HEMTs

Elison de Nazareth Matioli, Minghua Zhu, Riyaz Mohammed Abdul Khadar, Armin Jafari, Mohammad Samizadeh Nikooytabalvandani

GaN-on-Si high-electron-mobility transistors (HEMTs) exhibit excellent properties for efficient power conversion. Nevertheless, a considerable energy loss associated with the charging and discharging of the output capacitance (COSS) in these transistors se ...

Institute of Electrical and Electronics Engineers2021

Page 2 sur 2