arXiv Open Access 2024

Enabling full-speed random access to the entire memory on the A100 GPU

Alden Walker

Lihat Sumber

Abstrak

We describe some features of the A100 memory architecture. In particular, we give a technique to reverse-engineer some hardware layout information. Using this information, we show how to avoid TLB issues to obtain full-speed random HBM access to the entire memory, as long as we constrain any particular thread to a reduced access window of less than 64GB.

Topik & Kata Kunci

cs.PF cs.AR

Penulis (1)

Alden Walker

Format Sitasi

APA MLA BibTeX

Walker, A. (2024). Enabling full-speed random access to the entire memory on the A100 GPU. https://arxiv.org/abs/2405.11425

Akses Cepat

Lihat di Sumber

Informasi Jurnal

Tahun Terbit: 2024
Bahasa: en
Sumber Database: arXiv
Akses: Open Access ✓