Related papers: Generation of the Single Precision BLAS library fo…

Convolutional Neural Networks (CNNs) are widely used in deep learning applications, e.g. visual systems, robotics etc. However, existing software solutions are not efficient. Therefore, many hardware accelerators have been proposed…

In the world of linear algebra computation, a well-established standard exists called BLAS(Basic Linear Algebra Subprograms). This standard has been crucial for the development of software using linear algebra operations. Its benefits…

Modern SoC-FPGA that consists of FPGA with embedded ARM cores is being popularized as an embedded vision system platform. However, the design approach of SoC-FPGA applications still follows traditional hardware-software separate workflow,…

The reduction of a banded matrix to bidiagonal form is a critical step in the calculation of Singular Values, a cornerstone of scientific computing and AI. Although inherently parallel, this step has traditionally been considered unsuitable…

Matrix-matrix multiplication is a key computational kernel for numerous applications in science and engineering, with ample parallelism and data locality that lends itself well to high-performance implementations. Many matrix…

We present a novel, hardware-agnostic implementation strategy for lattice Boltzmann (LB) simulations, which yields massive performance on homogeneous and heterogeneous many-core platforms. Based solely on C++17 Parallel Algorithms, our…

Simulations based on particle methods, such as Smoothed Particle Hydrodynamics (SPH), are known to be computationally demanding. While such methods have for long been executed in parallel on multi-core CPUs, in recent years the increasing…

Stencil computation is one of the fundamental computing patterns in many application domains such as scientific computing and image processing. While there are promising studies that accelerate stencils on FPGAs, there lacks an automated…

RAPID-LLM is a unified performance modeling framework for large language model (LLM) training and inference on GPU clusters. It couples a DeepFlow-based frontend that generates hardware-aware, operator-level Chakra execution traces from an…

Multi-GPU programming traditionally requires developers to navigate complex trade-offs between performance and programmability. High-performance implementations typically rely on low-level HIP/CUDA communication libraries that demand…

The translation of linear algebra computations into efficient sequences of library calls is a non-trivial task that requires expertise in both linear algebra and high-performance computing. Almost all high-level languages and libraries for…

This study explores the use of automatic BLAS offloading and INT8-based emulation for accelerating traditional HPC workloads on modern GPU architectures. Through the use of low-bitwidth integer units and cache-coherent Unified Memory…

To increase performance and efficiency, systems use FPGAs as reconfigurable accelerators. A key challenge in designing these systems is partitioning computation between processors and an FPGA. An appropriate division of labor may be…

Sparse General Matrix Multiply (SpGEMM) is key for various High-Performance Computing (HPC) applications such as genomics and graph analytics. Using the semiring abstraction, many algorithms can be formulated as SpGEMM, allowing…

We present FooPar, an extension for highly efficient Parallel Computing in the multi-paradigm programming language Scala. Scala offers concise and clean syntax and integrates functional programming features. Our framework FooPar combines…

The Simplex tableau has been broadly used and investigated in the industry and academia. With the advent of the big data era, ever larger problems are posed to be solved in ever larger machines whose architecture type did not exist in the…

Dynamic programming (DP) based algorithms are essential yet compute-intensive parts of numerous bioinformatics pipelines, which typically involve populating a 2-D scoring matrix based on a recursive formula, optionally followed by a…

In this work, we present a new approach to high level synthesis (HLS), where high level functions are first mapped to an architectural template, before hardware synthesis is performed. As FPGA platforms are especially suitable for…

Deploying a large language model (LLM) inference service remains costly because centralized serving depends on specialized GPU clusters and high-bandwidth interconnects in datacenters. An appealing alternative is to leverage collaborative…

The rapid advancement of AI workloads and domain-specific architectures has led to increasingly diverse processor microarchitectures, whose design exploration requires fast and accurate performance validation. However, traditional workflows…