Publications related to Uniform memory access

EdgeAI-Aware Design of In-Memory Computing Architectures

Driven by the demand for real-time processing and the need to minimize latency in AI algorithms, edge computing has experienced remarkable progress. Decision-making AI applications stand out for their heavy reliance on data-centric operations, predominantl ...

EPFL2024

Chaosity: Understanding Contemporary NUMA-architectures

Anastasia Ailamaki, Viktor Sanca, Syed Mohammad Aunn Raza, Hamish Mcniece Hill Nicholson, Andreea Nica

Modern hardware is increasingly complex, requiring increasing effort to understand in order to carefully engineer systems for optimal performance and effective utilization. Moreover, established design principles and assumptions are not portable to modern ...

2023

Rebooting Virtual Memory with Midgard

Siddharth Gupta

Virtual Memory (VM) is a critical programming abstraction that is widely used in various modern computing platforms. With the rise of datacenter computing and birth of planet-scale online services, the semantic and capacity requirements from memory have ev ...

EPFL2023

Latency Interfaces for Systems Code

Rishabh Ramesh Iyer

This thesis demonstrates that it is feasible for systems code to expose a latency interface that describes its latency and related side effects for all inputs, just like the code's semantic interface describes its functionality and related side effects. S ...

EPFL2023

Decentralized in-order execution of a sequential task-based code for shared-memory architectures

Charly Nicolas Lucien Castes

The hardware complexity of modern machines makes the design of adequate programming models crucial for jointly ensuring performance, portability, and productivity in high-performance computing (HPC). Sequential task-based programming models paired with adv ...

IEEE COMPUTER SOC2022

Adaptive table placement in NUMA architectures

Anastasia Ailamaki, Iraklis Psaroudakis

Techniques and solutions are provided for performing adaptive database table placement in a non-uniform memory access (NUMA) architecture. The adaptive table placement can occur in response to changing workloads on the NUMA nodes. For example, if a particu ...

2022

An Accuracy-Driven Compression Methodology to Derive Efficient Codebook-Based CNNs

David Atienza Alonso, Giovanni Ansaloni, Miguel Peon Quiros, Flavio Ponzina

Codebook-based optimizations are a class of algorithmic-level transformations able to effectively reduce the computing and memory requirements of Convolutional Neural Networks (CNNs). This approach tightly limits the number of unique weights in each layer, ...

2022

Leaderless Consensus

Rachid Guerraoui, Vincent Gramoli, Mihail Igor Zablotchi, Karolos Antoniadis, Antoine Philippe Matthieu Desjardins

Classical synchronous consensus algorithms are leaderless: processes exchange their proposals, retain the maximum value and decide when they see the same choice across a couple of rounds. Indulgent consensus algorithms are more robust in that they only req ...

IEEE COMPUTER SOC2021

High-Level Synthesis of Dynamically Scheduled Circuits

Lana Josipovic

High-Level Synthesis (HLS) tools generate hardware designs from high-level programming languages. These tools almost universally build datapaths that are controlled using a centralized controller which relies on a static, compile-time schedule to determine ...

EPFL2021

Miss-Optimized Memory Systems: Turning Thousands of Outstanding Misses into Reuse Opportunities

Mikhail Asiatici

Even if Dennard scaling came to an end fifteen years ago, Moore' s law kept fueling an exponential growth in compute performance through increased parallelization. However, the performance of memory and, in particular, Dynamic Random Access Memory (DRAM), ...

EPFL2021

Using an antenna as a tool for studying microturbulence and zonal structures in tokamaks with a global gyrokinetic GPU-enabled particle-in-cell code

Noé Thomas Elie Ohana

In order to get a better understanding of the interaction of plasma microinstabilities and associated turbulence with specific modes, an antenna is implemented in the global gyrokinetic Particle-In-Cell (PIC) code ORB5. It consists in applying an external ...

EPFL2020

Quantum ESPRESSO toward the exascale

Nicola Marzari, Iurii Timrov, Roberto Car, Andrea Ferretti

Quantum ESPRESSO is an open-source distribution of computer codes for quantum-mechanical materials modeling, based on density-functional theory, pseudopotentials, and plane waves, and renowned for its performance on a wide range of hardware architectures, ...

AMER INST PHYSICS2020

Access pattern based optimization of memory access

Anastasia Ailamaki, Georgios Psaropoulos

A method for optimizing memory access for database operations is provided. The method may include identifying an access pattern associated with a database operation. The access pattern may include data required to perform the database operation. One or mor ...

2020

Heterogeneous Memory Organizations in Embedded Systems. Placement of Dynamic Data Objects

Miguel Peon Quiros

This book defines and explores the problem of placing the instances of dynamic data types on the components of the heterogeneous memory organization of an embedded system, with the final goal of reducing energy consumption and improving performance. It is ...

Springer International Publishing2020

Functionality Enhanced Memories for Edge-AI Embedded Systems

David Atienza Alonso, Alexandre Sébastien Julien Levisse, Pierre-Emmanuel Julien Marc Gaillardon, Marco Antonio Rios, William Andrew Simon

With the surge in complexity of edge workloads, it appeared in the scientific community that such workloads cannot be anymore overflown to the cloud due to the huge edge device to server communication energy cost and the high energy consumption induced in ...

IEEE2019

DynaBurst: Dynamically Assemblying DRAM Bursts over a Multitude of Random Accesses

Paolo Ienne, Mikhail Asiatici

The effective bandwidth of the FPGA external memory, usually DRAM, is extremely sensitive to the access pattern. Nonblocking caches that handle thousands of outstanding misses (miss-optimized memory systems) can dynamically improve bandwidth utilization wh ...

IEEE2019

Composite Data Types in Dynamic Dataflow Languages as Copyless Memory Sharing Mechanism

Marco Mattavelli, Endri Bezati, Aurélien François Gilbert Bloch

This paper presents new optimization approaches aiming at reducing the impact of memory accesses on the performance of dataflow programs. The approach is based on introducing a high level management of composite data types in dynamic dataflow programming l ...

2019

A Product Engine for Energy-Efficient Execution of Binary Neural Networks Using Resistive Memories

David Atienza Alonso, Marina Zapater Sancho, Pierre-Emmanuel Julien Marc Gaillardon, Yasir Mahmood Qureshi, Xifan Tang, Edouard Giacomin, João Miguel Morgado Pereira Vieira

The need for running complex Machine Learning (ML) algorithms, such as Convolutional Neural Networks (CNNs), in edge devices, which are highly constrained in terms of computing power and energy, makes it important to execute such applications efficiently. ...

2019

In Search of Lost Bandwidth: Extensive Reordering of DRAM Accesses on FPGA

Paolo Ienne, Mikhail Asiatici, Gabor Andras Csordas

For efficient acceleration on FPGA, it is essential for external memory to match the throughput of the processing pipelines. However, the usable DRAM bandwidth decreases significantly if the access pattern causes frequent row conflicts. Memory controllers ...

IEEE COMPUTER SOC2019

Pythia: Remote Oracles for the Masses

Mathias Josef Payer, Yuqing Zhang

Remote Direct Memory Access (RDMA) is a technology that allows direct access from the network to a machine's main memory without involving its CPU. RDMA offers low-latency, high-bandwidth performance and low CPU utilization. While RDMA provides massive per ...

USENIX ASSOC2019