Related papers: Adding MFMA Support to gem5

Mixture-of-Experts (MoE) model architectures can significantly reduce the number of activated parameters per token, enabling computationally efficient training and inference. However, their large overall parameter counts and model sizes…

Mixture of Experts (MoE) models have enabled the scaling of Large Language Models (LLMs) and Vision Language Models (VLMs) by achieving massive parameter counts while maintaining computational efficiency. However, MoEs introduce several…

In this new version of ZMCintegral, we have added the functionality of multi-function integrations, i.e. the ability to integrate more than $10^{3}$ different functions on GPUs. The Python API remains the similar as the previous versions.…

Graphics Processing Units (GPUs) are widely-used accelerators for data-parallel applications. In many GPU applications, GPU memory bandwidth bottlenecks performance, causing underutilization of GPU cores. Hence, disabling many cores does…

In recommendation systems, practitioners observed that increase in the number of embedding tables and their sizes often leads to significant improvement in model performances. Given this and the business importance of these models to major…

While Mixture of Experts (MoE) models achieve remarkable efficiency by activating only subsets of parameters, they suffer from high memory access costs during inference. Memory-layer architectures offer an appealing alternative with very…

We evaluate IBM's Enhanced Cell Broadband Engine (BE) as a possible building block of a new generation of lattice QCD machines. The Enhanced Cell BE will provide full support of double-precision floating-point arithmetics, including…

Many recent computational accelerators provide non-standard (e.g., reduced precision) arithmetic operations to enhance performance for floating-point matrix multiplication. Unfortunately, the properties of these accelerators are not widely…

Convolutional Neural Networks (CNNs) serve various applications with diverse performance and resource requirements. Model-aware CNN accelerators best address these diverse requirements. These accelerators usually combine multiple dedicated…

Usage of GPUs as co-processors is a well-established approach to accelerate costly algorithms operating on matrices and vectors. We aim to further improve the performance of the Global Neutrino Analysis framework (GNA) by adding GPU support…

Mixture-of-Expert (MoE) models enable efficient inference by employing smaller experts and activating only a subset of them per token. MoE serving engines distribute experts across multiple GPUs and route tokens to appropriate GPUs at…

Self-supervised pretraining has transformed computer vision by enabling data-efficient fine-tuning, yet high-resolution training typically requires server-scale infrastructure, limiting in-domain foundation model development for many…

Quantization significantly accelerates inference in large language models (LLMs) by replacing original high-precision matrices with low-precision counterparts. Recent advances in weight-activation quantization have primarily focused on…

Multi-access Edge Computing (MEC) is expected to act as the enabler for the integration of 5G (and future 6G) communication technologies with cloud-computing-based capabilities at the edge of the network. This will enable low-latency and…

NVIDIA Tensor Cores and AMD Matrix Cores (together called Matrix Accelerators) are of growing interest in high-performance computing and machine learning owing to their high performance. Unfortunately, their numerical behaviors are not…

Faster, cheaper, and more power efficient optimization solvers than those currently offered by general-purpose solutions are required for extending the use of model predictive control (MPC) to resource-constrained embedded platforms. We…

We approach the problem of implementing mixed-datatype support within the general matrix multiplication (GEMM) operation of the BLIS framework, whereby each matrix operand A, B, and C may be stored as single- or double-precision real or…

The rapid growth of large language models (LLMs) has driven the need for high-performance, scalable GPU hardware capable of efficiently serving models with hundreds of billions of parameters. While NVIDIA GPUs have traditionally dominated…

General matrix-matrix multiplication (GEMM) is a cornerstone of AI computations, making tensor processing engines (TPEs) increasingly critical in GPUs and domain-specific architectures. Existing architectures primarily optimize dataflow or…

Mobile Edge Computing (MEC) holds excellent potential in Congestion Management (CM) of 6G vehicular networks. A reasonable schedule of MEC ensures a more reliable and efficient CM system. Unfortunately, existing parallel and sequential…