Spatten: Efficient sparse attention architecture with cascade token and head pruning
The attention mechanism is becoming increasingly popular in Natural Language Processing
(NLP) applications, showing superior performance than convolutional and recurrent …
(NLP) applications, showing superior performance than convolutional and recurrent …
Dsagen: Synthesizing programmable spatial accelerators
Domain-specific hardware accelerators can provide orders of magnitude speedup and
energy efficiency over general purpose processors. However, they require extensive manual …
energy efficiency over general purpose processors. However, they require extensive manual …
Fifer: Practical acceleration of irregular applications on reconfigurable architectures
Coarse-grain reconfigurable arrays (CGRAs) can achieve much higher performance and
efficiency than general-purpose cores, approaching the performance of a specialized design …
efficiency than general-purpose cores, approaching the performance of a specialized design …
Snafu: an ultra-low-power, energy-minimal cgra-generation framework and architecture
Ultra-low-power (ULP) devices are becoming pervasive, enabling many emerging sensing
applications. Energy-efficiency is paramount in these applications, as efficiency determines …
applications. Energy-efficiency is paramount in these applications, as efficiency determines …
Recent developments in low-power AI accelerators: A survey
As machine learning and AI continue to rapidly develop, and with the ever-closer end of
Moore's law, new avenues and novel ideas in architecture design are being created and …
Moore's law, new avenues and novel ideas in architecture design are being created and …
OverGen: Improving FPGA usability through domain-specific overlay generation
FPGAs have been proven to be powerful computational accelerators across many types of
workloads. The mainstream programming approach is high level synthesis (HLS), which …
workloads. The mainstream programming approach is high level synthesis (HLS), which …
Riptide: A programmable, energy-minimal dataflow compiler and architecture
Emerging sensing applications create an unprecedented need for energy efficiency in
programmable processors. To achieve useful multi-year deployments on a small battery or …
programmable processors. To achieve useful multi-year deployments on a small battery or …
OpenCGRA: An open-source unified framework for modeling, testing, and evaluating CGRAs
Coarse-grained reconfigurable arrays (CGRAs), loosely defined as arrays of functional units
(eg, adder, subtractor, multiplier, divider, or larger multi-operation units, but smaller than a …
(eg, adder, subtractor, multiplier, divider, or larger multi-operation units, but smaller than a …
Lisa: Graph neural network based portable mapping on spatial accelerators
Spatial accelerators, such as Coarse-Grained Reconfigurable Arrays (CGRA), provide a
promising pathway to scale the performance and power efficiency of computing systems …
promising pathway to scale the performance and power efficiency of computing systems …
Energy efficient design of coarse-grained reconfigurable architectures: Insights, trends and challenges
E Aliagha, D Göhringer - 2022 International Conference on …, 2022 - ieeexplore.ieee.org
Coarse-Grained Reconfigurable Architectures (CGRAs) are promising solutions to achieve
more performance with the end of Moore's law. Thanks to word-level programmability, they …
more performance with the end of Moore's law. Thanks to word-level programmability, they …