Pub Date : 2020-06-01DOI: 10.1109/dsn-s50200.2020.00046
Dongning Ma, Xun Jiao
Various approximation methods demonstrate the effectiveness of voltage scaling in digital circuits in order to explore the energy-error trade-off. An accurate error model is of critical importance for assessing the error behavior of voltage-scaled circuits and its effects on the application quality. However, existing error models of voltage-scaled circuits overlook the effects of input data and error rate disparity among different bits. To tackle this challenge, we propose a machine learning-based error model of voltage-scaled circuits that can predict the timing error rate for each output bit. We train this model using random forest methods with input features and output labels extracted from gate-level simulation. We evaluate the model accuracy on different circuits. Across all bit positions, voltage levels, and circuits, the model achieves on average a relative error of 1.06%. The model also achieves an average per-voltage Root Mean Square Error (RMSE) of 0.92% and per-bit RMSE of 1.02%. Exposing this error rate even up to the application level, the model can estimate the quality of an image processing application under voltage scaling with an average accuracy of 97.5%.
{"title":"A Machine Learning-Based Error Model of Voltage-Scaled Circuits","authors":"Dongning Ma, Xun Jiao","doi":"10.1109/dsn-s50200.2020.00046","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00046","url":null,"abstract":"Various approximation methods demonstrate the effectiveness of voltage scaling in digital circuits in order to explore the energy-error trade-off. An accurate error model is of critical importance for assessing the error behavior of voltage-scaled circuits and its effects on the application quality. However, existing error models of voltage-scaled circuits overlook the effects of input data and error rate disparity among different bits. To tackle this challenge, we propose a machine learning-based error model of voltage-scaled circuits that can predict the timing error rate for each output bit. We train this model using random forest methods with input features and output labels extracted from gate-level simulation. We evaluate the model accuracy on different circuits. Across all bit positions, voltage levels, and circuits, the model achieves on average a relative error of 1.06%. The model also achieves an average per-voltage Root Mean Square Error (RMSE) of 0.92% and per-bit RMSE of 1.02%. Exposing this error rate even up to the application level, the model can estimate the quality of an image processing application under voltage scaling with an average accuracy of 97.5%.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"134 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122055024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-s50200.2020.00038
Ligeng Chen
Closed-source programs lack crucial information vital for code analysis because that information is stripped on compilation to achieve smaller executable size. Variable type information is fundamental in this process. In this paper, we implement a system called CATI (Context-Assisted Type Inference), which locates variables from stripped binaries and infers 19 types from variables. Experiments show that it infers variable type with 71.2% accuracy on unseen binaries.
{"title":"What Exactly Determines the Type? Inferring Types with Context","authors":"Ligeng Chen","doi":"10.1109/dsn-s50200.2020.00038","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00038","url":null,"abstract":"Closed-source programs lack crucial information vital for code analysis because that information is stripped on compilation to achieve smaller executable size. Variable type information is fundamental in this process. In this paper, we implement a system called CATI (Context-Assisted Type Inference), which locates variables from stripped binaries and infers 19 types from variables. Experiments show that it infers variable type with 71.2% accuracy on unseen binaries.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123955512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-s50200.2020.00009
S. Bagchi, François Taïani
range of from large-scale distributed systems and blockchains, to machine learning and security, cross-layer resilience tutorials delivered world-class a
从大规模分布式系统和区块链，到机器学习和安全，跨层弹性教程提供了世界一流的解决方案
{"title":"Message from the Tutorials Chairs","authors":"S. Bagchi, François Taïani","doi":"10.1109/dsn-s50200.2020.00009","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00009","url":null,"abstract":"range of from large-scale distributed systems and blockchains, to machine learning and security, cross-layer resilience tutorials delivered world-class a","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"62 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125881186","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-s50200.2020.00016
Fred Lin, A. Davoli, I. Akbar, Sukumar Kalmanje, Leandro Silva, J. Stamford, Yanai S. Golany, Jim Piazza, S. Sankar
Large-scale service environments rely on autonomous systems for remediating hardware failures efficiently. In production, the autonomous system diagnoses hardware failures based on the rules that the subject matter experts put in the system. This process is increasingly complex given new types of failures and the increasing complexity in the hardware and software configurations. In this paper, we present a machine learning framework that predicts the required remediations for undiagnosed failures, based on the similar repair tickets closed in the past. We explain the methodology in detail for setting up a machine learning model, deploying it in a production environment, and monitoring its performance with the necessary metrics. We also demonstrate the prediction performance on some of the repair actions.
{"title":"Predicting Remediations for Hardware Failures in Large-Scale Datacenters","authors":"Fred Lin, A. Davoli, I. Akbar, Sukumar Kalmanje, Leandro Silva, J. Stamford, Yanai S. Golany, Jim Piazza, S. Sankar","doi":"10.1109/dsn-s50200.2020.00016","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00016","url":null,"abstract":"Large-scale service environments rely on autonomous systems for remediating hardware failures efficiently. In production, the autonomous system diagnoses hardware failures based on the rules that the subject matter experts put in the system. This process is increasingly complex given new types of failures and the increasing complexity in the hardware and software configurations. In this paper, we present a machine learning framework that predicts the required remediations for undiagnosed failures, based on the similar repair tickets closed in the past. We explain the methodology in detail for setting up a machine learning model, deploying it in a production environment, and monitoring its performance with the necessary metrics. We also demonstrate the prediction performance on some of the repair actions.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127645040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-s50200.2020.00039
Paulo Silva
Given the large adoption and economical impact of permissionless blockchains, the complexity of the underlying systems and the adversarial environment in which they operate, it is fundamental to properly study and understand the emergent behavior and properties of these systems. We describe our experience on a detailed, one-month study of the Ethereum network from several geographically dispersed observation points. We leverage multiple geographic vantage points to assess the key pillars of Ethereum, namely geographical dispersion, network efficiency, blockchain efficiency and security, and the impact of mining pools. Among other new findings, we identify previously undocumented forms of selfish behavior and show that the prevalence of powerful mining pools exacerbates the geographical impact on block propagation delays. Furthermore, we provide a set of open measurement and processing tools, as well as the data set of the collected measurements, in order to promote further research on understanding permissionless blockchains.
{"title":"Impact of Geo-Distribution and Mining Pools on Blockchains: A Study of Ethereum - Practical Experience Report and Ongoing PhD Work","authors":"Paulo Silva","doi":"10.1109/dsn-s50200.2020.00039","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00039","url":null,"abstract":"Given the large adoption and economical impact of permissionless blockchains, the complexity of the underlying systems and the adversarial environment in which they operate, it is fundamental to properly study and understand the emergent behavior and properties of these systems. We describe our experience on a detailed, one-month study of the Ethereum network from several geographically dispersed observation points. We leverage multiple geographic vantage points to assess the key pillars of Ethereum, namely geographical dispersion, network efficiency, blockchain efficiency and security, and the impact of mining pools. Among other new findings, we identify previously undocumented forms of selfish behavior and show that the prevalence of powerful mining pools exacerbates the geographical impact on block propagation delays. Furthermore, we provide a set of open measurement and processing tools, as well as the data set of the collected measurements, in order to promote further research on understanding permissionless blockchains.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127665490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-S50200.2020.00020
Nathan D. Mickulicz, P. Narasimhan
Large-scale, high-density Wi-Fi networks use hundreds of access points to serve thousands of closely-packed users within a large physical space, such as within a stadium or arena. It is difficult to predict when and where problems will occur in these Wi-Fi networks, due to the constant movement of mobile devices within the network and the constantly-changing workload as users switch between applications. In this paper, we describe a unique approach to detecting, diagnosing, and mitigating problems in Wi-Fi networks using Wi-Fi performance data collected from mobile devices and shared between nearby peers. Our approach draws upon 3 years of production performance data that we have collected from 35 production mobile applications used in 25 professional and collegiate sports venues in the US. We also present an evaluation of the effectiveness of our diagnostic and mitigation approach in a real-world high-density Wi-Fi environment, showing that our approach outperforms standard driver-based problem detection and mitigation on several common Wi-Fi faults.
{"title":"Performance-Aware Wi-Fi Problem Diagnosis and Mitigation through Peer-to-Peer Data Sharing","authors":"Nathan D. Mickulicz, P. Narasimhan","doi":"10.1109/DSN-S50200.2020.00020","DOIUrl":"https://doi.org/10.1109/DSN-S50200.2020.00020","url":null,"abstract":"Large-scale, high-density Wi-Fi networks use hundreds of access points to serve thousands of closely-packed users within a large physical space, such as within a stadium or arena. It is difficult to predict when and where problems will occur in these Wi-Fi networks, due to the constant movement of mobile devices within the network and the constantly-changing workload as users switch between applications. In this paper, we describe a unique approach to detecting, diagnosing, and mitigating problems in Wi-Fi networks using Wi-Fi performance data collected from mobile devices and shared between nearby peers. Our approach draws upon 3 years of production performance data that we have collected from 35 production mobile applications used in 25 professional and collegiate sports venues in the US. We also present an evaluation of the effectiveness of our diagnostic and mitigation approach in a real-world high-density Wi-Fi environment, showing that our approach outperforms standard driver-based problem detection and mitigation on several common Wi-Fi faults.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128989648","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-S50200.2020.00013
Deborah S. Katz, Milda Zizyte, Casidhe Hutchison, David Guttendorf, Patrick E. Lanigan, Eric Sample, P. Koopman, Michael D. Wagner, Claire Le Goues
Robustness testing is an important technique to reveal defects and vulnerabilities in software, especially software for Unmanned Autonomous Systems (UAS). We present Robustness Inside Out Testing (RIOT) as a technique directed at finding failures in autonomy systems that are able to be activated from external interfaces. The technique consists of four main steps: unit-level robustness testing, generalization, permeability analysis, and activation. Each of these steps yields a valuable deliverable in the testing process, and, when applied in succession, expands a unit-level bug to an external interface. RIOT has the following advantages over traditional robustness testing: it finds faults faster, it can find faults missed by traditional approaches, it identifies faults that can be triggered from inputs at an external interface, and it produces useful artifacts to aid in fault diagnosis and repair. In this paper, we outline each step of the RIOT process and provide an example of RIOT finding a bug on a real system that would not have been discovered using existing techniques.
{"title":"Robustness Inside Out Testing","authors":"Deborah S. Katz, Milda Zizyte, Casidhe Hutchison, David Guttendorf, Patrick E. Lanigan, Eric Sample, P. Koopman, Michael D. Wagner, Claire Le Goues","doi":"10.1109/DSN-S50200.2020.00013","DOIUrl":"https://doi.org/10.1109/DSN-S50200.2020.00013","url":null,"abstract":"Robustness testing is an important technique to reveal defects and vulnerabilities in software, especially software for Unmanned Autonomous Systems (UAS). We present Robustness Inside Out Testing (RIOT) as a technique directed at finding failures in autonomy systems that are able to be activated from external interfaces. The technique consists of four main steps: unit-level robustness testing, generalization, permeability analysis, and activation. Each of these steps yields a valuable deliverable in the testing process, and, when applied in succession, expands a unit-level bug to an external interface. RIOT has the following advantages over traditional robustness testing: it finds faults faster, it can find faults missed by traditional approaches, it identifies faults that can be triggered from inputs at an external interface, and it produces useful artifacts to aid in fault diagnosis and repair. In this paper, we outline each step of the RIOT process and provide an example of RIOT finding a bug on a real system that would not have been discovered using existing techniques.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130111023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/dsn-s50200.2020.00033
Kshitiz Goel, Abhishek Bhaumick, D. Kaushal, S. Bagchi
Edge computing is actively being adopted by various organizations and applications owing to its bandwidth saving and faster response times. However, this is accompanied by its own set of reliability issues and serves as an excellent target for optimizations and analysis. Our work analyzes the effect of mobile clients on task failure rates and proposes a low overhead location and network congestion aware optimization. In this paper, we discuss our motivations, provide details about the dataset, present some statistical analysis, and propose an improved mobile-side edge selection policy.
{"title":"Reliability Analysis of Edge Scenarios Using Pedestrian Mobility","authors":"Kshitiz Goel, Abhishek Bhaumick, D. Kaushal, S. Bagchi","doi":"10.1109/dsn-s50200.2020.00033","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00033","url":null,"abstract":"Edge computing is actively being adopted by various organizations and applications owing to its bandwidth saving and faster response times. However, this is accompanied by its own set of reliability issues and serves as an excellent target for optimizations and analysis. Our work analyzes the effect of mobile clients on task failure rates and proposes a low overhead location and network congestion aware optimization. In this paper, we discuss our motivations, provide details about the dataset, present some statistical analysis, and propose an improved mobile-side edge selection policy.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122108404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-01DOI: 10.1109/DSN-S50200.2020.00018
Changchang Liu, Wei-Han Lee, S. Calo
In this work, we develop a new data transformation technique to mediate privacy-preserving access to data while achieving machine learning (ML) tasks. Specifically, we first leverage mutual information in information theory to quantify the utility-providing information (corresponding to any ML task) and the privacy information (could be arbitrary information specified by the users). We further convert the optimization of utility-privacy tradeoff into training a novel neural network (named as NeuralTran) which consists of three modules: transformation module, utility module and privacy module. NeuralTran can be leveraged to automatically transform the input data to ensure that only utility-providing information is kept while the private information is removed. Through extensive experiments on real world datasets, we show the effectiveness of NeuralTran in balancing utility and privacy as well as its advantages over previous approaches.
{"title":"Neuraltran: Optimal Data Transformation for Privacy-Preserving Machine Learning by Leveraging Neural Networks","authors":"Changchang Liu, Wei-Han Lee, S. Calo","doi":"10.1109/DSN-S50200.2020.00018","DOIUrl":"https://doi.org/10.1109/DSN-S50200.2020.00018","url":null,"abstract":"In this work, we develop a new data transformation technique to mediate privacy-preserving access to data while achieving machine learning (ML) tasks. Specifically, we first leverage mutual information in information theory to quantify the utility-providing information (corresponding to any ML task) and the privacy information (could be arbitrary information specified by the users). We further convert the optimization of utility-privacy tradeoff into training a novel neural network (named as NeuralTran) which consists of three modules: transformation module, utility module and privacy module. NeuralTran can be leveraged to automatically transform the input data to ensure that only utility-providing information is kept while the private information is removed. Through extensive experiments on real world datasets, we show the effectiveness of NeuralTran in balancing utility and privacy as well as its advantages over previous approaches.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133253519","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-05-05DOI: 10.1109/dsn-s50200.2020.00029
Martín Barrère, C. Hankin
In this paper, we present a novel MaxSAT-based technique to compute Maximum Probability Minimal Cut Sets (MPMCSs) in fault trees. We model the MPMCS problem as a Weighted Partial MaxSAT problem and solve it using a parallel SAT-solving architecture. The results obtained with our open source tool indicate that the approach is effective and efficient.
{"title":"Fault Tree Analysis: Identifying Maximum Probability Minimal Cut Sets with MaxSAT","authors":"Martín Barrère, C. Hankin","doi":"10.1109/dsn-s50200.2020.00029","DOIUrl":"https://doi.org/10.1109/dsn-s50200.2020.00029","url":null,"abstract":"In this paper, we present a novel MaxSAT-based technique to compute Maximum Probability Minimal Cut Sets (MPMCSs) in fault trees. We model the MPMCS problem as a Weighted Partial MaxSAT problem and solve it using a parallel SAT-solving architecture. The results obtained with our open source tool indicate that the approach is effective and efficient.","PeriodicalId":419045,"journal":{"name":"2020 50th Annual IEEE-IFIP International Conference on Dependable Systems and Networks-Supplemental Volume (DSN-S)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-05-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130173204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：