2018 · 88 citations · 18 references
ComputingComputer EngineeringComputer ScienceArtificial IntelligenceComputer ArchitectureComputer NetworksAccelerator ArchitectureAccelerator IntegrationHardware AccelerationProcessor MicroarchitectureAccelerator MicroarchitectureAcceleratorsLatency EffectsNetwork Interface CardsNetwork-on-chipProcessor Co-designProgrammable AcceleratorsProgrammable Networking
Artificial Neural Networks (NNs) play an increasingly important role in many services and applications, contributing significantly to compute infrastructures' workloads. When used in latency sensitive services, NNs are usually processed by CPUs since using an external dedicated hardware accelerator would be inefficient. However, with growing workloads size and complexity, CPUs are hitting their computation limits, requiring the introduction of new specialized hardware accelerators tailored to the task. In this paper we analyze the option to use programmable network devices, such as Network Cards and Switches, as NN accelerators in place of purpose built dedicated hardware. To this end, in this preliminary work we analyze in depth the properties of NN processing on CPUs, derive options to efficiently split such processing, and show that programmable network devices may be a suitable engine for implementing a CPU's NN co-processor.
18
In-Datacenter Performance Analysis of a Tensor Processing Unit
Norman P. Jouppi, Cliff Young, Nishant Patil et al. · 2017 · 4.4K citations
Pat Bosshart, Dan Daly, Glen Gibb et al. · ACM SIGCOMM Computer Communication Review · 2014 · 2.8K citations
In-Datacenter Performance Analysis of a Tensor Processing Unit
Norman P. Jouppi, Cliff Young, Nishant Patil et al. · ACM SIGARCH Computer Architecture News · 2017 · 1.3K citations · Full text
Itay Hubara, Daniel Soudry · arXiv (Cornell University) · 2016 · 924 citations · Full text