Salvato in:
Dettagli Bibliografici
Autori principali: Yi, Xiaoling, Antonio, Ryan, Deng, Yunhao, Kong, Fanchen, Dumoulin, Joren, Yin, Jun, Verhelst, Marian
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2602.11357
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866918333462347776
author Yi, Xiaoling
Antonio, Ryan
Deng, Yunhao
Kong, Fanchen
Dumoulin, Joren
Yin, Jun
Verhelst, Marian
author_facet Yi, Xiaoling
Antonio, Ryan
Deng, Yunhao
Kong, Fanchen
Dumoulin, Joren
Yin, Jun
Verhelst, Marian
contents Achieving high compute utilization across a wide range of AI workloads is crucial for the efficiency of versatile DNN accelerators. This paper presents the Voltra chip and its utilization-optimised DNN accelerator architecture, which leverages 3-Dimensional (3D) spatial data reuse along with efficient and flexible shared memory access. The 3D spatial dataflow enables balanced spatial data reuse across three dimensions, improving spatial utilization by up to 2.0x compared to a conventional 2D design. Inside the shared memory access architecture, Voltra incorporates flexible data streamers that enable mixed-grained hardware data pre-fetching and dynamic memory allocation, further improving the temporal utilization by 2.12-2.94x and achieving 1.15-2.36x total latency speedup compared with the non-prefetching and separated memory architecture, respectively. Fabricated in 16nm technology, our chip achieves 1.60 TOPS/W peak system energy efficiency and 1.25 TOPS/mm2 system area efficiency, which is competitive with state-of-the-art solutions while achieving high utilization across diverse workloads.
format Preprint
id arxiv_https___arxiv_org_abs_2602_11357
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A 16 nm 1.60TOPS/W High Utilization DNN Accelerator with 3D Spatial Data Reuse and Efficient Shared Memory Access
Yi, Xiaoling
Antonio, Ryan
Deng, Yunhao
Kong, Fanchen
Dumoulin, Joren
Yin, Jun
Verhelst, Marian
Hardware Architecture
Achieving high compute utilization across a wide range of AI workloads is crucial for the efficiency of versatile DNN accelerators. This paper presents the Voltra chip and its utilization-optimised DNN accelerator architecture, which leverages 3-Dimensional (3D) spatial data reuse along with efficient and flexible shared memory access. The 3D spatial dataflow enables balanced spatial data reuse across three dimensions, improving spatial utilization by up to 2.0x compared to a conventional 2D design. Inside the shared memory access architecture, Voltra incorporates flexible data streamers that enable mixed-grained hardware data pre-fetching and dynamic memory allocation, further improving the temporal utilization by 2.12-2.94x and achieving 1.15-2.36x total latency speedup compared with the non-prefetching and separated memory architecture, respectively. Fabricated in 16nm technology, our chip achieves 1.60 TOPS/W peak system energy efficiency and 1.25 TOPS/mm2 system area efficiency, which is competitive with state-of-the-art solutions while achieving high utilization across diverse workloads.
title A 16 nm 1.60TOPS/W High Utilization DNN Accelerator with 3D Spatial Data Reuse and Efficient Shared Memory Access
topic Hardware Architecture
url https://arxiv.org/abs/2602.11357