Though several deep learning (DL) detectors have been proposed for the network attack detection and achieved high accuracy, they are computationally expensive and struggle to satisfy the real-time detection for high-speed networks. Recently, programmable switches exhibit a remarkable throughput efficiency on production networks, indicating a possible deployment of the timely detector. Therefore, we present Soter, a DL enhanced in-network framework for the accurate real-time detection. Soter consists of two phases. One is filtering packets by a rule-based decision tree running on the Tofino ASIC. The other is executing a well-designed lightweight neural network for the thorough inspection of the suspicious packets on the CPU. Experiments on the commodity switch demonstrate that Soter behaves stably in ten network scenarios of different traffic rates and fulfills per-flow detection in 0.03s. Moreover, Soter naturally adapts to the distributed deployment among multiple switches, guaranteeing a higher total throughput for large data centers and cloud networks.
{"title":"Soter: Deep Learning Enhanced In-Network Attack Detection Based on Programmable Switches","authors":"Guorui Xie, Qing Li, Chupeng Cui, Peican Zhu, Dan Zhao, Wanxin Shi, Zhuyun Qi, Yong Jiang, Xianni Xiao","doi":"10.1109/SRDS55811.2022.00029","DOIUrl":"https://doi.org/10.1109/SRDS55811.2022.00029","url":null,"abstract":"Though several deep learning (DL) detectors have been proposed for the network attack detection and achieved high accuracy, they are computationally expensive and struggle to satisfy the real-time detection for high-speed networks. Recently, programmable switches exhibit a remarkable throughput efficiency on production networks, indicating a possible deployment of the timely detector. Therefore, we present Soter, a DL enhanced in-network framework for the accurate real-time detection. Soter consists of two phases. One is filtering packets by a rule-based decision tree running on the Tofino ASIC. The other is executing a well-designed lightweight neural network for the thorough inspection of the suspicious packets on the CPU. Experiments on the commodity switch demonstrate that Soter behaves stably in ten network scenarios of different traffic rates and fulfills per-flow detection in 0.03s. Moreover, Soter naturally adapts to the distributed deployment among multiple switches, guaranteeing a higher total throughput for large data centers and cloud networks.","PeriodicalId":143115,"journal":{"name":"2022 41st International Symposium on Reliable Distributed Systems (SRDS)","volume":"82 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128167777","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 : 2022-09-01DOI: 10.1109/SRDS55811.2022.00024
Zhihui Zhao, Yicheng Zeng, Jinfang Wang, Hong Li, Hongsong Zhu, Limin Sun
The object detection tasks based on edge computing have received great attention. A common concern hasn't been addressed is that edge may be unreliable and uploads the incorrect data to cloud. Existing works focus on the consistency of the transmitted data by edge. However, in cases when the inputs and the outputs are inherently different, the authenticity of data processing has not been addressed. In this paper, we first simply model the tampering detection. Then, bases on the feature insertion and game theory, the tampering detection and economic incentives mechanism (TDEI) is proposed. In tampering detection, terminal negotiates a set of features with cloud and inserts them into the raw data, after the cloud determines whether the results from edge contain the relevant information. The honesty incentives employs game theory to instill the distrust among different edges, preventing them from colluding and thwarting the tampering detection. Meanwhile, the subjectivity of nodes is also considered. TDEI distributes the tampering detection to all edges and realizes the self-detection of edge results. Experimental results based on the KITTI dataset, show that the accuracy of detection is 95% and 80%, when terminal's additional overhead is smaller than 30% for image and 20% for video, respectively. The interference ratios of TDEI to raw data are about 16% for video and 0% for image, respectively. Finally, we discuss the advantage and scalability of TDEI.
{"title":"Detection and Incentive: A Tampering Detection Mechanism for Object Detection in Edge Computing","authors":"Zhihui Zhao, Yicheng Zeng, Jinfang Wang, Hong Li, Hongsong Zhu, Limin Sun","doi":"10.1109/SRDS55811.2022.00024","DOIUrl":"https://doi.org/10.1109/SRDS55811.2022.00024","url":null,"abstract":"The object detection tasks based on edge computing have received great attention. A common concern hasn't been addressed is that edge may be unreliable and uploads the incorrect data to cloud. Existing works focus on the consistency of the transmitted data by edge. However, in cases when the inputs and the outputs are inherently different, the authenticity of data processing has not been addressed. In this paper, we first simply model the tampering detection. Then, bases on the feature insertion and game theory, the tampering detection and economic incentives mechanism (TDEI) is proposed. In tampering detection, terminal negotiates a set of features with cloud and inserts them into the raw data, after the cloud determines whether the results from edge contain the relevant information. The honesty incentives employs game theory to instill the distrust among different edges, preventing them from colluding and thwarting the tampering detection. Meanwhile, the subjectivity of nodes is also considered. TDEI distributes the tampering detection to all edges and realizes the self-detection of edge results. Experimental results based on the KITTI dataset, show that the accuracy of detection is 95% and 80%, when terminal's additional overhead is smaller than 30% for image and 20% for video, respectively. The interference ratios of TDEI to raw data are about 16% for video and 0% for image, respectively. Finally, we discuss the advantage and scalability of TDEI.","PeriodicalId":143115,"journal":{"name":"2022 41st International Symposium on Reliable Distributed Systems (SRDS)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131245736","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 : 2022-07-13DOI: 10.1109/SRDS55811.2022.00021
Marc Frei, Jonghoon Kwon, Seyedali Tabaeiaghdaei, Marc Wyss, C. Lenzen, A. Perrig
Many critical computing applications rely on secure and dependable time which is reliably synchronized across large distributed systems. Today's time synchronization architectures are commonly based on global navigation satellite systems at the considerable risk of being exposed to outages, malfunction, or attacks against availability and accuracy. This paper describes a practical instantiation of a new global, Byzantine fault-tolerant clock synchronization approach that does not place trust in any single entity and is able to tolerate a fraction of faulty entities while still maintaining synchronization on a global scale among otherwise sovereign network topologies. Leveraging strong resilience and security properties provided by the path-aware SCION networking architecture, the presented design can be implemented as a backward compatible active standby solution for existing time synchronization deployments. Through extensive evaluation, we demonstrate that over 94 % of time servers reliably minimize the offset of their local clocks to real-time in the presence of up to 20 % malicious nodes, and all time servers remain synchronized with a skew of only 2 ms even after one year of reference clock outage.
{"title":"G-SINC: Global Synchronization Infrastructure for Network Clocks","authors":"Marc Frei, Jonghoon Kwon, Seyedali Tabaeiaghdaei, Marc Wyss, C. Lenzen, A. Perrig","doi":"10.1109/SRDS55811.2022.00021","DOIUrl":"https://doi.org/10.1109/SRDS55811.2022.00021","url":null,"abstract":"Many critical computing applications rely on secure and dependable time which is reliably synchronized across large distributed systems. Today's time synchronization architectures are commonly based on global navigation satellite systems at the considerable risk of being exposed to outages, malfunction, or attacks against availability and accuracy. This paper describes a practical instantiation of a new global, Byzantine fault-tolerant clock synchronization approach that does not place trust in any single entity and is able to tolerate a fraction of faulty entities while still maintaining synchronization on a global scale among otherwise sovereign network topologies. Leveraging strong resilience and security properties provided by the path-aware SCION networking architecture, the presented design can be implemented as a backward compatible active standby solution for existing time synchronization deployments. Through extensive evaluation, we demonstrate that over 94 % of time servers reliably minimize the offset of their local clocks to real-time in the presence of up to 20 % malicious nodes, and all time servers remain synchronized with a skew of only 2 ms even after one year of reference clock outage.","PeriodicalId":143115,"journal":{"name":"2022 41st International Symposium on Reliable Distributed Systems (SRDS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121317260","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 : 2022-05-04DOI: 10.1109/SRDS55811.2022.00022
Pedro Fouto, P. Costa, Nuno M. Preguiça, J. Leitao
Prototyping and implementing distributed algorithms, particularly those that address challenges related with fault-tolerance and dependability, is a time consuming task. This is, in part, due to the need of addressing low level aspects such as management of communication channels, controlling timeouts or periodic tasks, and dealing with concurrency issues. This has a significant impact for researchers that want to build prototypes for conducting experimental evaluation; practitioners that want to compare different design alternatives/solutions; and even for practical teaching activities on distributed algorithms courses. In this paper we present Babel, a novel framework to develop, implement, and execute distributed protocols and systems. Babel promotes an event driven programming and execution model that simplifies the task of translating typical specifications or descriptions of algorithms into performant prototypes, while allowing the programmer to focus on the relevant challenges of these algorithms by transparently handling time consuming low level aspects. Furthermore, Babel provides, and allows the definition of, networking components that can capture different network capabilities (e.g., P2P, Client/Server, p-accrual Failure Detector), making the code mostly independent from the underlying communication aspects. Babel was built to be generic and can be used to implement a wide variety of different classes of distributed protocols. We conduct our experimental work with two relevant case studies, a Peer-to-Peer application and a State Machine Replication application, that show the generality and ease of use of Babel and present competitive performance when compared with significantly more complex implementations.
{"title":"Babel: A Framework for Developing Performant and Dependable Distributed Protocols","authors":"Pedro Fouto, P. Costa, Nuno M. Preguiça, J. Leitao","doi":"10.1109/SRDS55811.2022.00022","DOIUrl":"https://doi.org/10.1109/SRDS55811.2022.00022","url":null,"abstract":"Prototyping and implementing distributed algorithms, particularly those that address challenges related with fault-tolerance and dependability, is a time consuming task. This is, in part, due to the need of addressing low level aspects such as management of communication channels, controlling timeouts or periodic tasks, and dealing with concurrency issues. This has a significant impact for researchers that want to build prototypes for conducting experimental evaluation; practitioners that want to compare different design alternatives/solutions; and even for practical teaching activities on distributed algorithms courses. In this paper we present Babel, a novel framework to develop, implement, and execute distributed protocols and systems. Babel promotes an event driven programming and execution model that simplifies the task of translating typical specifications or descriptions of algorithms into performant prototypes, while allowing the programmer to focus on the relevant challenges of these algorithms by transparently handling time consuming low level aspects. Furthermore, Babel provides, and allows the definition of, networking components that can capture different network capabilities (e.g., P2P, Client/Server, p-accrual Failure Detector), making the code mostly independent from the underlying communication aspects. Babel was built to be generic and can be used to implement a wide variety of different classes of distributed protocols. We conduct our experimental work with two relevant case studies, a Peer-to-Peer application and a State Machine Replication application, that show the generality and ease of use of Babel and present competitive performance when compared with significantly more complex implementations.","PeriodicalId":143115,"journal":{"name":"2022 41st International Symposium on Reliable Distributed Systems (SRDS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-05-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115045850","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 : 2022-04-28DOI: 10.1109/SRDS55811.2022.00012
Jin Xu, Chi Hong, Jiyue Huang, L. Chen, Jérémie Decouchant
Federated learning is a private-by-design distributed learning paradigm where clients train local models on their own data before a central server aggregates their local updates to compute a global model. Depending on the aggregation method used, the local updates are either the gradients or the weights of local learning models, e.g., FedAvg aggregates model weights. Unfortunately, recent reconstruction attacks apply a gradient inversion optimization on the gradient update of a single mini-batch to reconstruct the private data used by clients during training. As the state-of-the-art reconstruction attacks solely focus on single update, realistic adversarial scenarios are over-looked, such as observation across multiple updates and updates trained from multiple mini-batches. A few studies consider a more challenging adversarial scenario where only model updates based on multiple mini-batches are observable, and resort to computationally expensive simulation to untangle the underlying samples for each local step. In this paper, we propose AGIC, a novel Approximate Gradient Inversion Attack that efficiently and effectively reconstructs images from both model or gradient updates, and across multiple epochs. In a nutshell, AGIC (i) approximates gradient updates of used training samples from model updates to avoid costly simulation procedures, (ii) leverages gradient/model updates collected from multiple epochs, and (iii) assigns increasing weights to layers with respect to the neural network structure for reconstruction quality. We extensively evaluate AGIC on three datasets, namely CIFAR-10, CIFAR-100 and ImageNet. Our results show that AGIC increases the peak signal-to-noise ratio (PSNR) by up to 50% compared to two representative state-of-the-art gradient inversion attacks. Furthermore, AGIC is faster than the state-of-the-art simulation-based attack, e.g., it is 5x faster when attacking FedAvg with 8 local steps in between model updates.
{"title":"AGIC: Approximate Gradient Inversion Attack on Federated Learning","authors":"Jin Xu, Chi Hong, Jiyue Huang, L. Chen, Jérémie Decouchant","doi":"10.1109/SRDS55811.2022.00012","DOIUrl":"https://doi.org/10.1109/SRDS55811.2022.00012","url":null,"abstract":"Federated learning is a private-by-design distributed learning paradigm where clients train local models on their own data before a central server aggregates their local updates to compute a global model. Depending on the aggregation method used, the local updates are either the gradients or the weights of local learning models, e.g., FedAvg aggregates model weights. Unfortunately, recent reconstruction attacks apply a gradient inversion optimization on the gradient update of a single mini-batch to reconstruct the private data used by clients during training. As the state-of-the-art reconstruction attacks solely focus on single update, realistic adversarial scenarios are over-looked, such as observation across multiple updates and updates trained from multiple mini-batches. A few studies consider a more challenging adversarial scenario where only model updates based on multiple mini-batches are observable, and resort to computationally expensive simulation to untangle the underlying samples for each local step. In this paper, we propose AGIC, a novel Approximate Gradient Inversion Attack that efficiently and effectively reconstructs images from both model or gradient updates, and across multiple epochs. In a nutshell, AGIC (i) approximates gradient updates of used training samples from model updates to avoid costly simulation procedures, (ii) leverages gradient/model updates collected from multiple epochs, and (iii) assigns increasing weights to layers with respect to the neural network structure for reconstruction quality. We extensively evaluate AGIC on three datasets, namely CIFAR-10, CIFAR-100 and ImageNet. Our results show that AGIC increases the peak signal-to-noise ratio (PSNR) by up to 50% compared to two representative state-of-the-art gradient inversion attacks. Furthermore, AGIC is faster than the state-of-the-art simulation-based attack, e.g., it is 5x faster when attacking FedAvg with 8 local steps in between model updates.","PeriodicalId":143115,"journal":{"name":"2022 41st International Symposium on Reliable Distributed Systems (SRDS)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-04-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123097867","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 : 2021-04-15DOI: 10.1109/SRDS55811.2022.00011
A. Bellet, Anne-Marie Kermarrec, Erick Lavoie
The convergence speed of machine learning models trained with Federated Learning is significantly affected by heterogeneous data partitions, even more so in a fully decentralized setting without a central server. In this paper, we show that the impact of label distribution skew, an important type of data heterogeneity, can be significantly reduced by carefully designing the underlying communication topology. We present D-Cliques, a novel topology that reduces gradient bias by grouping nodes in sparsely interconnected cliques such that the label distribution in a clique is representative of the global label distribution. We also show how to adapt the updates of decentralized SGD to obtain unbiased gradients and implement an effective momentum with D-Cliques. Our extensive empirical evaluation on MNIST and CIFAR10 validates our design and demonstrates that our approach achieves similar convergence speed as a fully-connected topology, while providing a significant reduction in the number of edges and messages. In a 1000-node topology, D-Cliques require 98% less edges and 96% less total messages, with further possible gains using a small-world topology across cliques.
{"title":"D-Cliques: Compensating for Data Heterogeneity with Topology in Decentralized Federated Learning","authors":"A. Bellet, Anne-Marie Kermarrec, Erick Lavoie","doi":"10.1109/SRDS55811.2022.00011","DOIUrl":"https://doi.org/10.1109/SRDS55811.2022.00011","url":null,"abstract":"The convergence speed of machine learning models trained with Federated Learning is significantly affected by heterogeneous data partitions, even more so in a fully decentralized setting without a central server. In this paper, we show that the impact of label distribution skew, an important type of data heterogeneity, can be significantly reduced by carefully designing the underlying communication topology. We present D-Cliques, a novel topology that reduces gradient bias by grouping nodes in sparsely interconnected cliques such that the label distribution in a clique is representative of the global label distribution. We also show how to adapt the updates of decentralized SGD to obtain unbiased gradients and implement an effective momentum with D-Cliques. Our extensive empirical evaluation on MNIST and CIFAR10 validates our design and demonstrates that our approach achieves similar convergence speed as a fully-connected topology, while providing a significant reduction in the number of edges and messages. In a 1000-node topology, D-Cliques require 98% less edges and 96% less total messages, with further possible gains using a small-world topology across cliques.","PeriodicalId":143115,"journal":{"name":"2022 41st International Symposium on Reliable Distributed Systems (SRDS)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123348373","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：