CANOPY: A CNFET-based Process Variation Aware Systolic DNN Accelerator

CANOPY: A CNFET-based Process Variation Aware Systolic DNN Accelerator. Chu, C., Xu, D., Wang, Y., & <a href="https://homes.luddy.indiana.edu/fc7/" target="_bilank">Fan Chen</a> In ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED), 2022.

Paper doi abstract bibtex

Although systolic accelerators have become the dominant method for executing Deep Neural Networks (DNNs), their performance efficiency (quantified as Energy-Delay Product or EDP) is limited by the capabilities of silicon Field-Effect Transistors (FETs). FETs constructed from Carbon Nanotubes (CNTs) have demonstrated > 10 × EDP benefits, however, the processing variations inherent in carbon nanotube FETs (CNFETs) fabrication compromise the EDP benefits, resulting > 40% performance degradation. In this work, we study the impact of CNT process variations and present Canopy, a process variation aware systolic DNN accelerator by leveraging the spatial correlation in CNT variations. Canopy co-optimizes the architecture and dataflow to allow computing engines in a systolic array run at their best performance with non-uniform latency, minimizing the performance degradation incurred by CNT variations. Furthermore, we devise Canopy with dynamic reconfigurability such that the microarchitectural capability and its associated flexibility achieves an extra degree of adaptability with regard to the DNN topology and processing hyper-parameters (e.g., batch size). Experimental results show that Canopy improves the performance by 5.85 × (4.66 ×) and reduces the energy by 34% (90%) when inferencing a single (a batch of) input compared to the baseline design under an iso-area comparison across seven DNN workloads.

@inproceedings{2022ISLPED:CNT,
 author = {Cheng Chu and Dawen Xu and Ying Wang and {<a href="https://homes.luddy.indiana.edu/fc7/" target="_bilank">Fan Chen</a></span>}},
 title = {{CANOPY: A CNFET-based Process Variation Aware Systolic DNN Accelerator}},
 booktitle = {ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED)},
 year = {2022},
isbn = {9781450393546},
url = {https://doi.org/10.1145/3531437.3539703},
doi = {10.1145/3531437.3539703},
abstract = {Although systolic accelerators have become the dominant method for executing Deep Neural Networks (DNNs), their performance efficiency (quantified as Energy-Delay Product or EDP) is limited by the capabilities of silicon Field-Effect Transistors (FETs). FETs constructed from Carbon Nanotubes (CNTs) have demonstrated &gt; 10 \texttimes{} EDP benefits, however, the processing variations inherent in carbon nanotube FETs (CNFETs) fabrication compromise the EDP benefits, resulting &gt; 40% performance degradation. In this work, we study the impact of CNT process variations and present&nbsp;Canopy, a process variation aware systolic DNN accelerator by leveraging the spatial correlation in CNT variations. Canopy co-optimizes the architecture and dataflow to allow computing engines in a systolic array run at their best performance with non-uniform latency, minimizing the performance degradation incurred by CNT variations. Furthermore, we devise&nbsp;Canopy with dynamic reconfigurability such that the microarchitectural capability and its associated flexibility achieves an extra degree of adaptability with regard to the DNN topology and processing hyper-parameters (e.g., batch size). Experimental results show that&nbsp;Canopy improves the performance by 5.85 \texttimes{} (4.66 \texttimes{}) and reduces the energy by 34% (90%) when inferencing a single (a batch of) input compared to the baseline design under an iso-area comparison across seven DNN workloads.},
articleno = {24},
numpages = {6}
}

Downloads: 0

{"_id":"DLS2jbATX6dJwi5bE","bibbaseid":"chu-xu-wang-ahrefhttpshomesluddyindianaedufc7targetbilankfanchenaspan-canopyacnfetbasedprocessvariationawaresystolicdnnaccelerator-2022","author_short":["Chu, C.","Xu, D.","Wang, Y.","<a href=\"https://homes.luddy.indiana.edu/fc7/\" target=\"_bilank\">Fan Chen</a>"],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["Cheng"],"propositions":[],"lastnames":["Chu"],"suffixes":[]},{"firstnames":["Dawen"],"propositions":[],"lastnames":["Xu"],"suffixes":[]},{"firstnames":["Ying"],"propositions":[],"lastnames":["Wang"],"suffixes":[]},{"firstnames":[],"propositions":[],"lastnames":["<a href=\"https://homes.luddy.indiana.edu/fc7/\" target=\"_bilank\">Fan Chen</a>"],"suffixes":[]}],"title":"CANOPY: A CNFET-based Process Variation Aware Systolic DNN Accelerator","booktitle":"ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED)","year":"2022","isbn":"9781450393546","url":"https://doi.org/10.1145/3531437.3539703","doi":"10.1145/3531437.3539703","abstract":"Although systolic accelerators have become the dominant method for executing Deep Neural Networks (DNNs), their performance efficiency (quantified as Energy-Delay Product or EDP) is limited by the capabilities of silicon Field-Effect Transistors (FETs). FETs constructed from Carbon Nanotubes (CNTs) have demonstrated > 10 × EDP benefits, however, the processing variations inherent in carbon nanotube FETs (CNFETs) fabrication compromise the EDP benefits, resulting > 40% performance degradation. In this work, we study the impact of CNT process variations and present Canopy, a process variation aware systolic DNN accelerator by leveraging the spatial correlation in CNT variations. Canopy co-optimizes the architecture and dataflow to allow computing engines in a systolic array run at their best performance with non-uniform latency, minimizing the performance degradation incurred by CNT variations. Furthermore, we devise Canopy with dynamic reconfigurability such that the microarchitectural capability and its associated flexibility achieves an extra degree of adaptability with regard to the DNN topology and processing hyper-parameters (e.g., batch size). Experimental results show that Canopy improves the performance by 5.85 × (4.66 ×) and reduces the energy by 34% (90%) when inferencing a single (a batch of) input compared to the baseline design under an iso-area comparison across seven DNN workloads.","articleno":"24","numpages":"6","bibtex":"@inproceedings{2022ISLPED:CNT,\n author = {Cheng Chu and Dawen Xu and Ying Wang and {<a href=\"https://homes.luddy.indiana.edu/fc7/\" target=\"_bilank\">Fan Chen</a>}},\n title = {{CANOPY: A CNFET-based Process Variation Aware Systolic DNN Accelerator}},\n booktitle = {ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED)},\n year = {2022},\nisbn = {9781450393546},\nurl = {https://doi.org/10.1145/3531437.3539703},\ndoi = {10.1145/3531437.3539703},\nabstract = {Although systolic accelerators have become the dominant method for executing Deep Neural Networks (DNNs), their performance efficiency (quantified as Energy-Delay Product or EDP) is limited by the capabilities of silicon Field-Effect Transistors (FETs). FETs constructed from Carbon Nanotubes (CNTs) have demonstrated > 10 \\texttimes{} EDP benefits, however, the processing variations inherent in carbon nanotube FETs (CNFETs) fabrication compromise the EDP benefits, resulting > 40% performance degradation. In this work, we study the impact of CNT process variations and present Canopy, a process variation aware systolic DNN accelerator by leveraging the spatial correlation in CNT variations. Canopy co-optimizes the architecture and dataflow to allow computing engines in a systolic array run at their best performance with non-uniform latency, minimizing the performance degradation incurred by CNT variations. Furthermore, we devise Canopy with dynamic reconfigurability such that the microarchitectural capability and its associated flexibility achieves an extra degree of adaptability with regard to the DNN topology and processing hyper-parameters (e.g., batch size). Experimental results show that Canopy improves the performance by 5.85 \\texttimes{} (4.66 \\texttimes{}) and reduces the energy by 34% (90%) when inferencing a single (a batch of) input compared to the baseline design under an iso-area comparison across seven DNN workloads.},\narticleno = {24},\nnumpages = {6}\n}\n\n\n","author_short":["Chu, C.","Xu, D.","Wang, Y.","<a href=\"https://homes.luddy.indiana.edu/fc7/\" target=\"_bilank\">Fan Chen</a>"],"key":"2022ISLPED:CNT","id":"2022ISLPED:CNT","bibbaseid":"chu-xu-wang-ahrefhttpshomesluddyindianaedufc7targetbilankfanchenaspan-canopyacnfetbasedprocessvariationawaresystolicdnnaccelerator-2022","role":"author","urls":{"Paper":"https://doi.org/10.1145/3531437.3539703"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://homes.luddy.indiana.edu/fc7/fan-publication.bib","dataSources":["SzsArXknrJasD2EPT"],"keywords":[],"search_terms":["canopy","cnfet","based","process","variation","aware","systolic","dnn","accelerator","chu","xu","wang","<a href=\"https://homes.luddy.indiana.edu/fc7/\" target=\"_bilank\">fan chen</a>"],"title":"CANOPY: A CNFET-based Process Variation Aware Systolic DNN Accelerator","year":2022}