Related papers: AMReX and pyAMReX: Looking Beyond ECP
Robotics has made remarkable hardware strides-from DARPA's Urban and Robotics Challenges to the first humanoid-robot kickboxing tournament-yet commercial autonomy still lags behind progress in machine learning. A major bottleneck is…
The emergence of CXL (Compute Express Link) promises to transform the status of interconnects between host and devices and in turn impact the design of all software layers. With its low overhead, low latency, and memory coherency…
Several manufacturers have already started to commercialize near-bank Processing-In-Memory (PIM) architectures. Near-bank PIM architectures place simple cores close to DRAM banks and can yield significant performance and energy improvements…
We present EasyRec, an easy-to-use, extendable and efficient recommendation framework for building industrial recommendation systems. Our EasyRec framework is superior in the following aspects: first, EasyRec adopts a modular and pluggable…
Existing methods for merging experts during model training and fine-tuning predominantly rely on Euclidean geometry, which assumes a flat parameter space. This assumption can limit the model's generalization ability, especially during the…
Machine Learning (ML)-enabled systems challenge traditional Requirements Engineering (RE) and agile management due to data dependence, experimentation, and uncertain model behavior. Existing RE and agile practices remain poorly integrated…
To achieve scalability with today's heterogeneous HPC resources, we need a dramatic shift in our thinking; MPI+X is not enough. Asynchronous Many Task (AMT) runtime systems break down the global barriers imposed by the Bulk Synchronous…
Developing complex, reliable advanced accelerators requires a coordinated, extensible, and comprehensive approach in modeling, from source to the end of beam lifetime. We present highlights in Exascale Computing to scale accelerator…
Adaptive Mesh Refinement (AMR) enables efficient computation of flows by providing high resolution in critical regions while allowing for coarsening in areas where fine detail is unnecessary. While early AMR software packages relied solely…
Today, using multiple heterogeneous accelerators efficiently from applications and high-level frameworks, such as TensorFlow and Caffe, poses significant challenges in three respects: (a) sharing accelerators, (b) allocating available…
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 matrix-vector multiplications, which dominate these…
In this work, we revisit the marking decisions made in the standard adaptive finite element method (AFEM). Experience shows that a na\"{i}ve marking policy leads to inefficient use of computational resources for adaptive mesh refinement…
We propose a framework for performing differentiable geometric modeling based on the Function Representation (FRep). The framework is built on top of modern libraries for performing automatic differentiation allowing us to obtain…
Multi-access Edge Computing (MEC) will enable context-aware services for users of mobile 4G/5G networks. MEC application developers need tools to aid the design and the performance evaluation of their apps. During the early stages of…
Several applications in astrophysics require adequately resolving many physical and temporal scales which vary over several orders of magnitude. Adaptive mesh refinement techniques address this problem effectively but often result in…
We present the current status of the apeNEXT project. Aim of this project is the development of the next generation of APE machines which will provide multi-teraflop computing power. Like previous machines, apeNEXT is based on a custom…
This report presents the design of the Scope infrastructure for extensible and portable benchmarking. Improvements in high- performance computing systems rely on coordination across different levels of system abstraction. Developing and…
We introduce TyXe, a Bayesian neural network library built on top of Pytorch and Pyro. Our leading design principle is to cleanly separate architecture, prior, inference and likelihood specification, allowing for a flexible workflow where…
Tomographic imaging has benefited from advances in X-ray sources, detectors and optics to enable novel observations in science, engineering and medicine. These advances have come with a dramatic increase of input data in the form of faster…
Understanding the nonlinear dynamics of coupled scalar fields often necessitates simulations on a 3D mesh. These simulations can be computationally expensive if a large scale separation is involved. A common solution is adaptive mesh…