Software Passports for Automated Performance Anomaly Detection of Cyber-Physical Systems

Odyurt, Uraz; Meyer, Hugo; Pimentel, Andy D.; Paradas, E.; Alonso, Ignacio González

doi:10.1007/978-3-030-27562-4_18

Cited by 6 publications

(10 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the notions of software passports and software signatures apply to systems with repetitive nature, here we refer to an execution as a bounded segment of the overall operation of the system. These bounded segments, phases, can be in different granularities, namely atomic phases and combo-phases [18].…”

Section: Software Passports and Signaturesmentioning

confidence: 99%

“…Simulation output traces corresponding to baseline executions are used to create software passports, intended to act as a signature construct for runtime performance behaviour of reference executions. During system operation, software signatures are dynamically created with the purpose of detecting abnormal performance behaviour of processes that can cause system-level performance anomalies [18]. As will be explained later on, regression modelling is used to create software signatures and passports for comparative purposes.…”

Section: An Analytics-based Approach To Self-adaptive Cpsmentioning

confidence: 99%

“…To create our software passports, reference executions are used and patched traces are simulated, after which the output is processed to extract performance related values such as, CPU utilisation, read and write counts over time, per process and per execution phase [18]. Different regression models are evaluated considering different goodness of fit tests, and the best regression function is selected to represent each performance metric for each process in a phase.…”

Section: Predictive Stage In Detailmentioning

confidence: 99%

“…Our fault injection implementation focuses on affecting the duration of events by modifying the end-time of events [18]. The software allows us to completely configure the percentage of processes to be affected, the percentage of events to be affected, the types of events (communication or computation), the type of performance anomaly to be introduced (transient or persistent), as well as the overlap between the processes selected to be modified, introducing transient or persistent anomalies.…”

Section: Fault Injection Mechanismsmentioning

confidence: 99%

“…The creation of software passports, as explained in [18], considers a workload to be divided in different, but repetitive execution phases. In this set of experiments, we have considered one execution phase that corresponds to a repetitive task performed by an ASML semiconductor photolithography system.…”

Section: Software Passports and Signaturesmentioning

confidence: 99%

See 4 more Smart Citations

An analytics-based method for performance anomaly classification in cyber-physical systems

Meyer

Odyurt

Pimentel

et al. 2020

Proceedings of the 35th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

The analysis and correct categorisation of software performance anomalies is a major challenge in current industrial Cyber-Physical Systems (CPS). The automated evaluation of runtime performance metrics provides an overview of the software behaviour of CPS and may allow discovering or even predicting the occurrence of software performance anomalies. We present an approach to automatically identify deviations from expected performance behaviour of software processes running in a distributed industrial CPS. Our approach consists of collecting process performance signatures, using regression modelling techniques. This involves determining per process signatures and signature violations, enabling system anomaly type detection and classification. We evaluate different classification algorithms and predict the type of system performance anomaly that a signature violation can provoke. We demonstrate in our experiment that our design is capable of detecting and classifying synthetically introduced performance anomalies to real execution tracing data from a real semiconductor photolithography machine. Initial results show that we can achieve up to 93% of accuracy when classifying performance anomalies. CCS CONCEPTS• Computer systems organization → Embedded and cyberphysical systems; Reliability; Availability;

show abstract

Section: Software Passports and Signaturesmentioning

confidence: 99%

Section: An Analytics-based Approach To Self-adaptive Cpsmentioning

confidence: 99%

Section: Predictive Stage In Detailmentioning

confidence: 99%

Section: Fault Injection Mechanismsmentioning

confidence: 99%

Section: Software Passports and Signaturesmentioning

confidence: 99%

See 3 more Smart Citations

An analytics-based method for performance anomaly classification in cyber-physical systems

Meyer

Odyurt

Pimentel

et al. 2020

Proceedings of the 35th Annual ACM Symposium on Applied Computing

Self Cite

View full text Add to dashboard Cite

show abstract

The Choice of AI Matters: Alternative Machine Learning Approaches for CPS Anomalies

Odyurt

Sapra

Pimentel

2021

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

We compare the pros and cons of two Artificial Intelligence (AI) solutions, addressing the anomaly detection and identification challenge in industrial Cyber-Physical Systems (CPS). We demonstrate how our current approach, Advanced DL, based on Convolutional Neural Networks (CNN) differs from a previous one, Classic ML. Though both workflows prove to result in highly accurate classification of anomalies, Classic ML is superior in this regard with 99.23% accuracy against 94.85%. This comes at a cost, as Classic ML requires total insight and expertise regarding the system under scrutiny and heavy amounts of feature engineering, while Advanced DL treats the data as a black box, minimising the effort. At the same time, we show that finding the best performing CNN model design is not trivial. We present a quantitative comparison of both workflows in terms of elapsed times for training, validation and preprocessing, alongside discussions on qualitative aspects. Such a comparison, involving analysis of workflows for the given use-case, is of independent interest. We find the choice of AI solution to be use-case dependent.

show abstract

Building Golden Signal Based Signatures for Log Anomaly Detection

Nagar

Samanta

Mohapatra

et al. 2022

2022 IEEE 15th International Conference on Cloud Computing (CLOUD)

View full text Add to dashboard Cite

Software Passports for Automated Performance Anomaly Detection of Cyber-Physical Systems

Cited by 6 publications

References 11 publications

An analytics-based method for performance anomaly classification in cyber-physical systems

An analytics-based method for performance anomaly classification in cyber-physical systems

The Choice of AI Matters: Alternative Machine Learning Approaches for CPS Anomalies

Building Golden Signal Based Signatures for Log Anomaly Detection

Contact Info

Product

Resources

About