Pasar al contenido principal

SOFTWARE DE SISTEMA Y HERRAMIENTAS DE NUEVA GENERACION PARA ARQUITECTURAS EMERGENTES

PID2024-158311NB-I00

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal para la Investigación y el Desarrollo Experimental
Subprograma Subprograma Estatal de Generación de Conocimiento Científico-Técnico y Desarrollo Experimental
Convocatoria Proyectos de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2024
Plan Estatal 2024-2027 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD COMPLUTENSE DE MADRID

Resultados relacionados

Resultados totales (Incluyendo duplicados): 2
Encontrada(s) 1 página(s)

Docta Complutense
Publicación Artículo científico (JournalArticle)

The RISC-V FPGA (RVfpga) teaching package

  • Chaver Martínez, Daniel Ángel
  • Harris, Sarah
  • Piñuel Moreno, Luis
  • Kindgren, Olof
  • Kakakhel, Zubair
  • Owen, Chris
  • Kravitz, Roy
  • Gómez Pérez, José Ignacio
  • Castro Rodríguez, Fernando
  • Olcoz Herrero, Katzalin
  • Villalba Moreno, Julio
  • Grinshpun, Alexander
  • Gabbay, Freddu
  • Seed, Luke
  • Duarte, Rui
  • López, Manuel
  • Alonso, Óscar
  • Owen, Robert
RISC-V is a free and open-standard ISA based on RISC principles, allowing anyone to design, manufacture, and sell RISC-V chips and software. Its flexibility and growing ecosystem have made it popular in research, education, and industry, increasing the need for educational materials. This paper provides an in-depth description of the RVfpga course, which offers a solid introduction to computer architecture using the RISC-V instruction set and FPGA technology. It focuses on providing hands-on experience with real-world RISC-V cores, the VeeR EH1 and EL2 cores, developed by Western Digital and hosted by ChipsAlliance. The course targets students and educators in computing-related fields, enabling them to integrate practical RISC-V knowledge into their curricula. The course materials, which include detailed labs, setup guides, and the full SoC source code in System Verilog, are available for free. Students learn to compile, debug, and run C and assembly programs, to interact with built-in peripherals, to extend the SoC, and to explore microarchitectural features.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

The Cambrian Explosion of Mixed-Precision Matrix Multiplication for Quantized Deep Learning Inference

  • Martínez, Héctor
  • Igual, Francisco D.
  • Castelló, Adrián|||0000-0002-8576-8451
  • Quintana-Ortí, Enrique S.|||0000-0002-5454-165X
[EN] Recent advances in deep learning (DL) have promoted to a shift from traditional 64-bit floating point (FP64) arithmetic for scientific computing toward reduced-precision formats-such as FP16, BF16, or even 8-bit integers-combined with mixed-precision arithmetic. This transition enhances computational throughput, reduces memory and bandwidth usage, and improves energy efficiency, offering significant advantages for resource-constrained edge devices. To support this shift, hardware architectures have evolved accordingly, now including adapted ISAs (Instruction Set Architectures) that expose mixed-precision vector units and matrix engines tailored for DL workloads. At the heart of many DL and scientific computing tasks is the general matrix-matrix multiplication (GEMM), a fundamental kernel historically optimized using fused multiply-add (FMA) vector instructions on SIMD (single instruction, multiple data) units. However, as hardware moves toward mixed-precision dot (or inner)-product-centric operations optimized for quantized inference, these legacy approaches are being phased out. In response to this, our paper revisits the conventional, high-performance implementation of GEMM and describes strategies for adapting it to mixed integer precision (MIP) arithmetic across modern ISAs, including x86_64, Arm, and RISC-V. Concretely, we illustrate novel micro-kernel designs and data layouts that better exploit today's specialized hardware and demonstrate significant performance gains from MIP arithmetic over floatingpoint implementations across three representative CPUs. These contributions highlight a new era of GEMM optimization-driven by the demands of DL inference on heterogeneous architectures, marking what we term as the "Cambrian period" for matrix multiplication., This work is supported by grants PID2021-126576NB-I00, PID2023-146569NB-C2 and PID2024-158311NB-I00, funded by MCIN/AEI/10.13039/5011000 11033, by "ERDF A way of making Europe", Programa PROMETEO2023 (CIPROM/2022/20) by Conselleria de Innovacio, Universidades, Ciencia y Sociedad Digital from the Generalitat Valenciana, and by Ayudaa Primeros Proyectosde Investigacion (PAID-06-24) , Vicerrectorado de Investigacion de la Universitat Politecnica de Valencia. A. Castello is supported by the CIAPOS/2023/431 grant, and Fondo Social Europeo Plus 2021-2027 from the Generalitat Valenciana.