An Approximation-based Fault Detection Scheme for Image Processing Applications

Biasielli, Matteo; Cassano, Luca; Miele, Antonio

doi:10.23919/date48585.2020.9116425

Cited by 7 publications

(4 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At the hardware level, other approaches have exploited adding approximate circuits to reduce the duplication overhead [7]- [9], [26], [27]. Work in [7] proposes to design a dedicated approximated hardware circuit for detecting errors in specific operations, while in [8], [9] authors design approximated checkers to catch just errors affecting the most significant portions of the mantissa or just the exponent.…”

Section: Related Work and Contributions To The State-of-the-artmentioning

confidence: 99%

Reduced Precision DWC: An Efficient Hardening Strategy for Mixed-Precision Architectures

Santos

Brandalero

Sullivan

et al. 2022

IEEE Trans. Comput.

View full text Add to dashboard Cite

Duplication with Comparison (DWC) is an effective software-level solution to improve the reliability of computing devices. However, it introduces performance and energy consumption overheads that could be unsuitable for high-performance computing or real-time safety-critical applications.In this work, we present Reduced-Precision Duplication with Comparison (RP-DWC) as a means to lower the overhead of DWC by executing the redundant copy in reduced precision. RP-DWC is particularly suitable for modern mixed-precision architectures, such as NVIDIA GPUs, that feature dedicated functional units for computing with programmable accuracy. We discuss the benefits and challenges associated with RP-DWC and show that the intrinsic difference between the mixed-precision copies allows for detecting most, but not all, errors. However, as the undetected faults are the ones that fall into the difference between precisions, they are the ones that produce a much smaller impact on the application output and, thus, might be tolerated.We investigate RP-DWC impact into fault detection, performance, and energy consumption on Volta GPUs. Through fault injection and beam experiment, using three microbenchmarks and four real applications, we show that RP-DWC achieves an excellent coverage (up to 86%) with minimal overheads (as low as 0.1% time and 24% energy consumption overhead)

show abstract

Section: Related Work and Contributions To The State-of-the-artmentioning

confidence: 99%

Reduced Precision DWC: An Efficient Hardening Strategy for Mixed-Precision Architectures

Santos

Brandalero

Sullivan

et al. 2022

IEEE Trans. Comput.

View full text Add to dashboard Cite

show abstract

“…However, this approach suffers from the classical 2x overhead caused by the two exact replicas inherited by the DWC. The work proposed in the current paper moves from [7] and, based on the preliminary results presented [14], proposes a new solution, based on AC to evaluate the entire image as usable or not, and at the same time to drastically reduce the execution time of the hardened application. Finally, it is worth mentioning the system-level fault detection scheme proposed in [24] where, similarly to [11], no replica is required; however, this scheme is tailored to a specific image processing algorithm and cannot be applied in general.…”

Section: Related Workmentioning

confidence: 99%

“…A preliminary version of the proposed approach has been presented in [14]; we here extend it to achieve a mature proposal that includes the following novel contributions:…”

Section: Introductionmentioning

confidence: 99%

Approximation-based Fault Tolerance in Image Processing Applications

Biasielli¹,

Bolchini

Cassano

et al. 2021

IEEE Trans. Emerg. Topics Comput.

Self Cite

View full text Add to dashboard Cite

Image processing applications exhibit an intrinsic degree of fault tolerance due to i) the redundant nature of images, and ii) the possible ability of the consumers of the application output to effectively carry out their task even when it is slightly corrupted. In this application scenario the classical Duplication with Comparison (DWC) scheme, that rejects images (and requires re-executions) when the two replicas' outputs differ in a per-pixel comparison, may be over-conservative. In this paper, we propose a novel lightweight fault tolerant scheme specifically tailored for image processing applications. The proposed scheme enhances the state-of-the-art by: i) improving the DWC scheme by replacing one of the two exact replicas with an approximated counterpart, and ii) allowing to distinguish between usable and unusable images instead of corrupted and uncorrupted ones by means of a Convolutional Neural Network-based checker.To tune the proposed scheme we introduce a specific design methodology that optimizes both execution time and fault detection capability of the hardened system. We report the results of the application of the proposed approach on two case studies; our proposal achieves an average execution time reduction larger than 30% w.r.t. the DWC with re-execution, and less than 4% misclassified unusable images.

show abstract

“…the end-user application. This approach has been later enhanced in [10] by approximating the replica, thus achieving a considerable performance improvement.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Lightweight Fault Detection and Management for Image Restoration

Bolchini

Cassano

Miele

et al. 2020

2020 IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems (DFT)

Self Cite

View full text Add to dashboard Cite

Image restoration is generally employed to recover an image that has been blurred, for example, for noise suppression purposes. The Richardson-Lucy (RL) algorithm is a widely used iterative approach for image restoration. In this paper we propose a lightweight application-specific fault detection and management scheme for RL that exploits two specific characteristics of such algorithm: i) there is a strong correlation between the input and output images of each iteration, and ii) the algorithm is often able to produce a final output that is very similar to the expected one although the output of an intermediate iteration has been corrupted by a fault. The proposed scheme exploits these characteristics to detect the occurrence of a fault without requiring duplication and to determine whether the error in the output of an intermediate iteration of the algorithm would be absorbed (thus avoiding image dropping and algorithm reexecution) or whether the image has to be discarded and the overall elaboration to be re-executed. An experimental campaign demonstrated that our scheme allows for an execution time reduction of about 54% w.r.t. the classical Duplication with Comparison (DWC), still providing about 99% fault detection.

show abstract

An Approximation-based Fault Detection Scheme for Image Processing Applications

Cited by 7 publications

References 12 publications

Reduced Precision DWC: An Efficient Hardening Strategy for Mixed-Precision Architectures

Reduced Precision DWC: An Efficient Hardening Strategy for Mixed-Precision Architectures

Approximation-based Fault Tolerance in Image Processing Applications

Lightweight Fault Detection and Management for Image Restoration

Contact Info

Product

Resources

About