Is this you? As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.
Claim your profile
Articles by Torsten Hoefler
Proximity is all You Need - Two Tricks for Chiplet Interconnects
By Torsten Hoefler, ETH Zürich As chiplet-based designs gain momentum, the challenge of optimizing topologies within technology and packaging constraints becomes ever more pressing. This talk explores two powerful techniques for improving chiplet interconnects developed under the lead of my student Patrick Iff while adhering to distance limitations imposed by various packaging technologies.
FoldedHexaTorus: An Inter-Chiplet Interconnect Topology for Chiplet-based Systems using Organic and Glass Substrates
By Patrick Iff, Maciej Besta and Torsten Hoefler ETH Zurich Chiplet-based systems are rapidly gaining traction in the market. Two packaging options for such systems are the established organic substrates and the emerging glass substrates. These substrates are used to implement the inter-chiplet interconnect (ICI), which is crucial for overall system performance. To guide the development of ICIs, we introduce three design principles for ICI network topologies on organic and glass substrates.
HammingMesh: A Network Topology for Large-Scale Deep Learning
Numerous microarchitectural optimizations unlocked tremendous processing power for deep neural networks that in turn fueled the AI revolution. With the exhaustion of such optimizations, the growth of modern AI is now gated by the performance of training systems, especially their data movement. Instead of focusing on single accelerators, we investigate data-movement characteristics of large-scale training at full system scale.
[2404.12953] Low-Depth Spatial Tree Algorithms
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Digital twins of Earth and the computing challenge of human interaction - Nature Computational Science
Digital twins of Earth have the capability to offer versatile access to detailed information on our changing world, helping societies to adapt to climate change and to manage the effects of local impacts, globally. Nevertheless, human interaction with digital twins requires advances in computational science, particularly where complex geophysical data is turned into information to support decision making.
Cppless: Productive and Performant Serverless Programming in C++
Happy π Day! Pi Day is finally here – and so is Giving Day for arXiv! Donate today to directly support arXiv initiatives and help keep open science open. Infinite decimals. Infinite new ideas to discover. Infinite reasons to give.
Disentangling Hype from Practicality: On Realistically Achieving Quantum Advantage
Home/Magazine Archive/May 2023 (Vol. 66, No. 5)/Disentangling Hype from Practicality: On Realistically.../Full Text Review Articles Disentangling Hype from Practicality: On Realistically Achieving Quantum Advantage By Torsten Hoefler, Thomas Häner, Matthias Troyer Communications of the ACM, May 2023, Vol.
Disentangling Hype from Practicality: On Realistically Achieving Quantum Advantage
By Torsten Hoefler, Thomas Häner, Matthias Troyer Communications of the ACM, May 2023, Vol.
[2301.02432] Myths and Legends in High-Performance Computing
In just 3 minutes help us understand how you see arXiv.
[2201.07598] Near-Optimal Sparse Allreduce for Distributed Deep Learning
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
The Future of Deep Learning Will be Sparse
By Torsten Hoefler, Dan Alistarh, Nikoli Dryden, and Tal Ben-Nun Deep learning continues to deliver surprising new capabilities for tasks such as image and object classification, game play, translation, and even molecular structure prediction and other significant advancements in the computational sciences. These tasks are often carried out at human or superhuman performance levels, and one could argue that machines take over where human understanding of complex systems ends.
The digital revolution of Earth-system science
Abstract Computational science is crucial for delivering reliable weather and climate predictions. However, despite decades of high-performance computing experience, there is serious concern about the sustainability of this application in the post-Moore/Dennard era. Here, we discuss the present limitations in the field and propose the design of a novel infrastructure that is scalable and more adaptable to future, yet unknown computing architectures.
[1805.08288] Transformations of High-Level Synthesis Codes for High-Performance Computing
Authors:Johannes de Fine Licht, Maciej Besta, Simon Meierhans, Torsten Hoefler (Submitted on 21 May 2018 (v1), last revised 29 Oct 2019 (this version, v5)) Abstract: Specialized hardware architectures promise a major step in performance and energy efficiency over the traditional load/store devices currently employed in large scale computing systems.
Enabling Highly Scalable Remote Memory Access Programming with MPI-3 One Sided
Modern high-performance networks offer remote direct memory access (RDMA) that exposes a process' virtual address space to other processes in the network. The Message Passing Interface (MPI) specification has recently been extended with a programming interface called MPI-3 Remote Memory Access (MPI-3 RMA) for efficiently exploiting state-of-the-art RDMA features. MPI-3 RMA enables a powerful programming model that alleviates many message passing downsides.
Demystifying Parallel and Distributed Deep Learning: An In-Depth Concurrency Analysis
; Computer Vision and Pattern Recognition (cs.CV); Distributed, Parallel, and Cluster Computing (cs.DC); Neural and Evolutionary Computing (cs.NE)
Show More
loading
Actions
Is this you?
As a journalist, you can create a free Muck Rack account to customize your profile, list your contact preferences, and upload a portfolio of your best work.Get in touch with Torsten
Contact Torsten, search articles and posts on X, monitor coverage, and track replies from one place.
Learn more about Muck Rack