Energy efficient boosting of GEMM accelerators for DNN via reuse

buir.contributor.authorÖzturk, Özcan
dc.citation.epage26en_US
dc.citation.issueNumber5en_US
dc.citation.spage1en_US
dc.citation.volumeNumber27en_US
dc.contributor.authorCicek, Nihat Mert
dc.contributor.authorShen, Xipeng
dc.contributor.authorÖzturk, Özcan
dc.date.accessioned2023-02-25T10:59:59Z
dc.date.available2023-02-25T10:59:59Z
dc.date.issued2022-06-06
dc.departmentDepartment of Computer Engineeringen_US
dc.description.abstractReuse-centric convolutional neural networks (CNN) acceleration speeds up CNN inference by reusing computations for similar neuron vectors in CNN’s input layer or activation maps. This new paradigm of optimizations is, however, largely limited by the overheads in neuron vector similarity detection, an important step in reuse-centric CNN. This article presents an in-depth exploration of architectural support for reuse-centric CNN. It addresses some major limitations of the state-of-the-art design and proposes a novel hardware accelerator that improves neuron vector similarity detection and reduces the energy consumption of reuse-centric CNN inference. The accelerator is implemented to support a wide variety of neural network settings with a banked memory subsystem. Design exploration is performed through RTL simulation and synthesis on an FPGA platform. When integrated into Eyeriss, the accelerator can potentially provide improvements up to 7.75× in performance. Furthermore, it can reduce the energy used for similarity detection up to 95.46%, and it can accelerate the convolutional layer up to 3.63× compared to the software-based implementation running on the CPU.en_US
dc.identifier.doi10.1145/3503469en_US
dc.identifier.eissn1557-7309en_US
dc.identifier.issn1084-4309en_US
dc.identifier.urihttp://hdl.handle.net/11693/111730en_US
dc.language.isoEnglishen_US
dc.publisherAssociation for Computing Machinery, Incen_US
dc.relation.isversionofhttps://dx.doi.org/10.1145/3503469en_US
dc.source.titleACM Transactions on Design Automation of Electronic Systemsen_US
dc.subjectReuseen_US
dc.subjectDeep neural networksen_US
dc.subjectGemmen_US
dc.subjectAcceleratoren_US
dc.titleEnergy efficient boosting of GEMM accelerators for DNN via reuseen_US
dc.typeArticleen_US

Files

Original bundle

Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
Energy_Efficient_Boosting_of_GEMM_Accelerators_for_DNN_via_Reuse.pdf
Size:
4.89 MB
Format:
Adobe Portable Document Format
Description:

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.69 KB
Format:
Item-specific license agreed upon to submission
Description: