Development of low-overhead soft error mitigation technique for safety critical neural networks applications

Deep Neural Networks (DNNs) have been widely applied in healthcare applications. DNN-based healthcare applications are safety-critical systems that require highreliability implementation due to a high risk of human death or injury in case of malfunction. Several DNN accelerators are used to execute...

Full description

Saved in:

Bibliographic Details
Main Author:	Khalid Adam, Ismail Hammad
Format:	Thesis
Language:	English
Published:	2021
Subjects:	T Technology (General)
Online Access:	http://umpir.ump.edu.my/id/eprint/34715/1/Development%20of%20low-overhead%20soft%20error%20mitigation%20technique%20for%20safety%20critical%20neural.ir.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-ump-ir.34715
record_format	uketd_dc
spelling	my-ump-ir.347152022-10-14T02:57:30Z Development of low-overhead soft error mitigation technique for safety critical neural networks applications 2021-05 Khalid Adam, Ismail Hammad T Technology (General) TA Engineering (General). Civil engineering (General) Deep Neural Networks (DNNs) have been widely applied in healthcare applications. DNN-based healthcare applications are safety-critical systems that require highreliability implementation due to a high risk of human death or injury in case of malfunction. Several DNN accelerators are used to execute these DNN models, and GPUs are currently the most prominent and the dominated DNN accelerators. However, GPUs are prone to soft errors that dramatically impact the GPU behaviors; such error may corrupt data values or logic operations, which result in Silent Data Corruption (SDC). The SDC propagates from the physical level to the application level (SDC that occurs in hardware GPUs’ components) results in misclassification of objects in DNN models, leading to disastrous consequences. Food and Drug Administration (FDA) reported that 1078 of the adverse events (10.1%) were unintended errors (i.e., soft errors) encountered, including 52 injuries and two deaths. Several traditional techniques have been proposed to protect electronic devices from soft errors by replicating the DNN models. However, these techniques cause significant overheads of area, performance, and energy, making them challenging to implement in healthcare systems that have strict deadlines. To address this issue, this study developed a Selective Mitigation Technique based on the standard Triple Modular Redundancy (S-MTTM-R) to determine the model’s vulnerable parts, distinguishing Malfunction and Light-Malfunction errors. A comprehensive vulnerability analysis was performed using a SASSIFI fault injector at the CNN AlexNet and DenseNet201 models: layers, kernels, and instructions to show both models’ resilience and identify the most vulnerable portions and harden them by injecting them while implemented on NVIDIA’s GPUs. The experimental results showed that S-MTTM-R achieved a significant improvement in error masking. No-Malfunction have been improved from 54.90%, 67.85%, and 59.36% to 62.80%, 82.10%, and 80.76% in the three modes RF, IOA, and IOV, respectively for AlexNet. For DenseNet, NoMalfunction have been improved from 43.70%, 67.70%, and 54.68% to 59.90%, 84.75%, and 83.07% in the three modes RF, IOA, and IOV, respectively. Importantly, S-MTTMR decreased the percentage of errors that case misclassification (Malfunction) from 3.70% to 0.38% and 5.23% to 0.23%, for AlexNet and DenseNet, respectively. The performance analysis results showed that the S-MTTM-R achieved lower overhead compared to the well-known protection techniques: Algorithm-Based Fault Tolerance (ABFT), Double Modular Redundancy (DMR), and Triple Modular Redundancy (TMR). In light of these results, the study revealed strong evidence that the developed S-MTTMR was successfully mitigated the soft errors for the DNNs model on GPUs with lowoverheads in energy, performance, and area indicated a remarkable improvement in the healthcare domains’ model reliability. 2021-05 Thesis http://umpir.ump.edu.my/id/eprint/34715/ http://umpir.ump.edu.my/id/eprint/34715/1/Development%20of%20low-overhead%20soft%20error%20mitigation%20technique%20for%20safety%20critical%20neural.ir.pdf pdf en public phd doctoral Universiti Malaysia Pahang College of Engineering
institution	Universiti Malaysia Pahang Al-Sultan Abdullah
collection	UMPSA Institutional Repository
language	English
topic	T Technology (General) T Technology (General)
spellingShingle	T Technology (General) T Technology (General) Khalid Adam, Ismail Hammad Development of low-overhead soft error mitigation technique for safety critical neural networks applications
description	Deep Neural Networks (DNNs) have been widely applied in healthcare applications. DNN-based healthcare applications are safety-critical systems that require highreliability implementation due to a high risk of human death or injury in case of malfunction. Several DNN accelerators are used to execute these DNN models, and GPUs are currently the most prominent and the dominated DNN accelerators. However, GPUs are prone to soft errors that dramatically impact the GPU behaviors; such error may corrupt data values or logic operations, which result in Silent Data Corruption (SDC). The SDC propagates from the physical level to the application level (SDC that occurs in hardware GPUs’ components) results in misclassification of objects in DNN models, leading to disastrous consequences. Food and Drug Administration (FDA) reported that 1078 of the adverse events (10.1%) were unintended errors (i.e., soft errors) encountered, including 52 injuries and two deaths. Several traditional techniques have been proposed to protect electronic devices from soft errors by replicating the DNN models. However, these techniques cause significant overheads of area, performance, and energy, making them challenging to implement in healthcare systems that have strict deadlines. To address this issue, this study developed a Selective Mitigation Technique based on the standard Triple Modular Redundancy (S-MTTM-R) to determine the model’s vulnerable parts, distinguishing Malfunction and Light-Malfunction errors. A comprehensive vulnerability analysis was performed using a SASSIFI fault injector at the CNN AlexNet and DenseNet201 models: layers, kernels, and instructions to show both models’ resilience and identify the most vulnerable portions and harden them by injecting them while implemented on NVIDIA’s GPUs. The experimental results showed that S-MTTM-R achieved a significant improvement in error masking. No-Malfunction have been improved from 54.90%, 67.85%, and 59.36% to 62.80%, 82.10%, and 80.76% in the three modes RF, IOA, and IOV, respectively for AlexNet. For DenseNet, NoMalfunction have been improved from 43.70%, 67.70%, and 54.68% to 59.90%, 84.75%, and 83.07% in the three modes RF, IOA, and IOV, respectively. Importantly, S-MTTMR decreased the percentage of errors that case misclassification (Malfunction) from 3.70% to 0.38% and 5.23% to 0.23%, for AlexNet and DenseNet, respectively. The performance analysis results showed that the S-MTTM-R achieved lower overhead compared to the well-known protection techniques: Algorithm-Based Fault Tolerance (ABFT), Double Modular Redundancy (DMR), and Triple Modular Redundancy (TMR). In light of these results, the study revealed strong evidence that the developed S-MTTMR was successfully mitigated the soft errors for the DNNs model on GPUs with lowoverheads in energy, performance, and area indicated a remarkable improvement in the healthcare domains’ model reliability.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Khalid Adam, Ismail Hammad
author_facet	Khalid Adam, Ismail Hammad
author_sort	Khalid Adam, Ismail Hammad
title	Development of low-overhead soft error mitigation technique for safety critical neural networks applications
title_short	Development of low-overhead soft error mitigation technique for safety critical neural networks applications
title_full	Development of low-overhead soft error mitigation technique for safety critical neural networks applications
title_fullStr	Development of low-overhead soft error mitigation technique for safety critical neural networks applications
title_full_unstemmed	Development of low-overhead soft error mitigation technique for safety critical neural networks applications
title_sort	development of low-overhead soft error mitigation technique for safety critical neural networks applications
granting_institution	Universiti Malaysia Pahang
granting_department	College of Engineering
publishDate	2021
url	http://umpir.ump.edu.my/id/eprint/34715/1/Development%20of%20low-overhead%20soft%20error%20mitigation%20technique%20for%20safety%20critical%20neural.ir.pdf
_version_	1783732207571959808

Development of low-overhead soft error mitigation technique for safety critical neural networks applications

Similar Items