Hardware and Software Support for Mixed Precision Computing: a Roadmap for Embedded and HPC Systems

Fornaciari, William; Agosta, Giovanni; Cattaneo, Daniele; Denisov, L N; Galimberti, Andrea; Magnani, G.; Zoni, Davide

doi:10.23919/date56975.2023.10137092

Cited by 2 publications

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

FPGA-Based Design and Implementation of a Code-Based Post-quantum KEM

Galimberti

2024

SpringerBriefs in Applied Sciences and Technology

View full text Add to dashboard Cite

Post-quantum cryptography aims to design cryptosystems that can be deployed on traditional computers and resist attacks from quantum computers, which are widely expected to break the currently deployed public-key cryptography solutions in the upcoming decades. Providing effective hardware support is crucial to ensuring a wide adoption of post-quantum cryptography solutions, and it is one of the requirements set by the USA’s National Institute of Standards and Technology within its ongoing standardization process. This research delivers a configurable FPGA-based hardware architecture to support BIKE, a post-quantum QC-MDPC code-based key encapsulation mechanism. The proposed architecture is configurable through a set of architectural and code parameters, which make it efficient, providing good performance while using the resources available on FPGAs effectively, flexible, allowing to support different large QC-MDPC codes defined by the designers of the cryptosystem, and scalable, targeting the whole Xilinx Artix-7 FPGA family. Two separate modules target the cryptographic functionality of the client and server nodes of the quantum-resistant key exchange, respectively, and a complexity-based heuristic that leverages the knowledge of the time and space complexity of the configurable hardware components steers the design space exploration to identify their best parameterization. The proposed architecture outperforms the state-of-the-art reference software that exploits the Intel AVX2 extension and runs on a desktop-class CPU by 1.77 and 1.98 times, respectively, for AES-128- and AES-192-equivalent security instances of BIKE, and it provides a speedup of more than six times compared to the fastest reference state-of-the-art hardware architecture, which targets the same FPGA family.

show abstract

FPGA-Based Design and Implementation of a Code-Based Post-quantum KEM

Galimberti

2024

SpringerBriefs in Applied Sciences and Technology

View full text Add to dashboard Cite

show abstract

Mixed-precision computing in the GRIST dynamical core for weather and climate modelling

Chen,

Zhang,

Wang

et al. 2024

Geosci. Model Dev.

View full text Add to dashboard Cite

Abstract. Atmosphere modelling applications are becoming increasingly memory-bound due to the inconsistent development rates between processor speeds and memory bandwidth. In this study, we mitigate memory bottlenecks and reduce the computational load of the Global–Regional Integrated Forecast System (GRIST) dynamical core by adopting a mixed-precision computing strategy. Guided by an application of the iterative development principle, we identify the coded equation terms that are precision insensitive and modify them from double to single precision. The results show that most precision-sensitive terms are predominantly linked to pressure gradient and gravity terms, while most precision-insensitive terms are advective terms. Without using more computing resources, computational time can be saved, and the physical performance of the model is largely kept. In the standard computational test, the reference runtime of the model's dry hydrostatic core, dry nonhydrostatic core, and the tracer transport module is reduced by 24 %, 27 %, and 44 %, respectively. A series of idealized tests, real-world weather and climate modelling tests, was performed to assess the optimized model performance qualitatively and quantitatively. In particular, in the long-term coarse-resolution climate simulation, the precision-induced sensitivity can manifest at the large scale, while in the kilometre-scale weather forecast simulation, the model's sensitivity to the precision level is mainly limited to small-scale features, and the wall-clock time is reduced by 25.5 % from the double- to mixed-precision full-model simulations.

show abstract

Hardware and Software Support for Mixed Precision Computing: a Roadmap for Embedded and HPC Systems

Cited by 2 publications

References 27 publications

FPGA-Based Design and Implementation of a Code-Based Post-quantum KEM

FPGA-Based Design and Implementation of a Code-Based Post-quantum KEM

Mixed-precision computing in the GRIST dynamical core for weather and climate modelling

Contact Info

Product

Resources

About