MUGNoC A Software-Configured Multicast-Unicast-Gather NoC for Accelerating CNN Dataflows

Current communication infrastructures for convolutional neural networks (CNNs) only focus on specific transmission patterns, not applicable to benefit the whole system if the dataflow changes or different dataflows run in one system. To reduce data movement, various CNN dataflows are presented. For...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2023 28th Asia and South Pacific Design Automation Conference (ASP-DAC) s. 308 - 313
Hlavní autoři: Chen, Hui, Liu, Di, Li, Shiqing, Huai, Shuo, Luo, Xiangzhong, Liu, Weichen
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: New York, NY, USA ACM 16.01.2023
Edice:ACM Conferences
Témata:
Boosting
CNN Dataflow
Design automation
Network-on-chips
Networks
Networks > Network algorithms
Networks > Network algorithms > Control path algorithms
Networks > Network algorithms > Control path algorithms > Network control algorithms
Networks > Network architectures
Networks > Network architectures > Network design principles
Networks > Network components
Networks > Network protocols
Networks > Network protocols > Network layer protocols
Networks > Network services
Networks > Network services > Programmable networks
Networks > Network types
Networks > Network types > Packet-switching networks
Parallel multipath transmission
Routing
Software
System performance
Traffic control
Unicast
parallel multipath transmission
CNN dataflow
network-on-chips
ISBN:9781450397834, 1450397832
ISSN:2153-697X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Current communication infrastructures for convolutional neural networks (CNNs) only focus on specific transmission patterns, not applicable to benefit the whole system if the dataflow changes or different dataflows run in one system. To reduce data movement, various CNN dataflows are presented. For these dataflows, parameters and results are delivered using different traffic patterns, i.e., multicast, unicast, and gather, preventing dataflow-specific communication backbones from benefiting the entire system if the dataflow changes or different dataflows run in the same system. Thus, in this paper, we propose MUG-NoC to support typical traffic patterns and accelerate them, therefore boosting multiple dataflows. Specifically, (i) we for the first time support multicast in 2D-mesh software configurable NoC by revising router configuration and proposing the efficient multicast routing; (ii) we decrease unicast latency by transmitting data through the different routes in parallel; (iii) we reduce output gather overheads by pipelining basic dataflow units. Experiments show that at least our proposed design can reduce 39.2% total data transmission time compared with the state-of-the-art CNN communication backbone.
ISBN:9781450397834
1450397832
ISSN:2153-697X
DOI:10.1145/3566097.3567846