2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)最新文献

{"title":"Stable Cuckoo Filter for Data Streams","authors":"Shangsen Li, Lailong Luo, Deke Guo, Yawei Zhao","doi":"10.1109/ICPADS53394.2021.00023","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00023","url":null,"abstract":"Cuckoo filter (CF), Bloom filter (BF) and their variants are space-efficient probabilistic data structures for approximate set membership queries. However, their data synopsis would inevitably become unusable when there are a number of member updates on the set; while updates are not uncommon for the real-world data streaming applications such as duplicate item detection, malicious URL checking, and caching applications. It has been shown that some variants of BF can be adaptive to stream applications. However, current extensions of BF structures generally incur unstable performance or intolerant membership testing errors. In this paper, we aim to design a data synopsis for membership testing on data streams with stable performance and tolerant query errors. To this end, we propose Stable Cuckoo Filters (SCF), which take a fine-grained manner to evict the stale elements and store those more recent ones. SCF absorbs the design philosophy from several unsuccessful designs. Specifically, SCFs take elegant update operations to embed time information with insertion operation and carefully evict the stale elements. We show that a tight upper bound of the expected false positive rate (FPR) remains asymptotically constant over the insertion of new members. The query error for recent elements of SCF (FNR) is related to the characteristics of the input data stream and query workloads. Extensive experiments on the real-world and synthetic datasets show that our designs are more stable than the existing variants of BF and realize 7 x smaller false errors and up to 3 x throughput.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125002861","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}

{"title":"ATO-EDGE: Adaptive Task Offloading for Deep Learning in Resource-Constrained Edge Computing Systems","authors":"Yihao Wang, Ling Gao, J. Ren, Rui Cao, Hai Wang, Jie Zheng, Quanli Gao","doi":"10.1109/ICPADS53394.2021.00025","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00025","url":null,"abstract":"On-device deep learning enables mobile devices to perform complex tasks, such as object detection and voice translation, regardless of the network condition. The advanced deep learning model gives an excellent performance, also leads to a heavy burden on resource-limited devices (i.e., mobile devices). To speed up the on-device deep learning. Prior studies focus on developing lightweight network architecture for real-time inference by sacrificing model accuracy. This paper presents ATO-EDGE: adaptive task offloading for deep learning based on edge computing. Considering three optimization goals, energy consumption, accuracy, and latency, ATO-EDGE leverages an offline pre-trained model to select a suitable deep learning model on a specific device to process the given task. We apply our approach to object detection and evaluate it on Jetson TX2, Xilinx ZYNQ 7020, and Raspberry 3B+. The deep learning model candidates contain ten typical object detection models trained on Microsoft COCO 2017 dataset. We obtain, on average, 28.25%, 35.44%, and 0.9 improvements respectively for latency, energy consumption, and mAP (mean average precision) when compared to the SOTA DETR model on the Raspberry Pi.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129322179","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}

{"title":"On Consensus Number 1 Objects","authors":"P. Khanchandani, Jan Schäppi, Ye Wang, Roger Wattenhofer","doi":"10.1109/ICPADS53394.2021.00115","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00115","url":null,"abstract":"The consensus number concept is used to determine the power of synchronization primitives in distributed systems. Recent work in the blockchain domain motivates shifting the attention to consensus number 1 objects, as it has been shown that transaction-based blockchains just need consensus number 1. In this paper we want to get a better understanding of such consensus number 1 objects. In particular, we study the necessary and sufficient conditions for determining the consensus number 1 objects. If an object has consensus number 1, then its operations must be either commutative or associative (necessary condition). On the other hand, if the operations are consistently commutative or overwriting, i.e., independent of the current state of the object, then the consensus number of the object is 1 (sufficient condition). We give an algorithm to implement such generic consensus number 1 objects using only read/write registers. This implies that read/write registers are universal enough to solve tasks, such as asset transfer of a cryptocurrency, among many others, in wait-free distributed systems for any number of processes.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123901322","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}

{"title":"Effective Anomaly Detection Based on Reinforcement Learning in Network Traffic Data","authors":"Zhongyang Wang, Yijie Wang, Hongzuo Xu, Yongjun Wang","doi":"10.1109/ICPADS53394.2021.00043","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00043","url":null,"abstract":"Mixed-type data with both categorical and numerical features are ubiquitous in network security, but the existing methods are minimal to deal with them. Existing methods usually process mixed-type data through feature conversion, whereas their performance is downgraded by information loss and noise caused by the transformation. Meanwhile, existing methods usually superimpose domain knowledge and machine learning in which fixed thresholds are used. It cannot dynamically adjust the anomaly threshold to the actual scenario, resulting in inaccurate anomalies obtained, which results in poor performance. To address these issues, this paper proposes a novel Anomaly Detection method based on Reinforcement Learning, termed ADRL, which uses reinforcement learning to dynamically search for thresholds and accurately obtain anomaly candidate sets, fusing domain knowledge and machine learning fully and promoting each other. Specifically, ADRL uses prior domain knowledge to label known anomalies and uses entropy and deep autoencoder in the categorical and numerical feature spaces, respectively, to obtain anomaly scores combining with known anomaly information, which are integrated to get the overall anomaly scores via a dynamic integration strategy. To obtain accurate anomaly candidate sets, ADRL uses reinforcement learning to search for the best threshold. Detailedly, it initializes the anomaly threshold to get the initial anomaly candidate set and carries on the frequent rule mining to the anomaly candidate set to form the new knowledge. Then, ADRL uses the obtained knowledge to adjust the anomaly score and get the score modification rate. According to the modification rate, different threshold modification strategies are executed, and the best threshold, that is, the threshold under the maximum modification rate, is finally obtained, and the modified anomaly scores are obtained. The scores are used to re-carry out machine learning to improve the algorithm's accuracy for anomalous data. Repeat the above process until the method is stable. We experiment on ten real network traffic datasets. Experiments show ADRL averagely improves ROC-AUC and PR-AUC than eight state-of-the-art competitors by 89.6% and 286.0%, respectively.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128081775","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}

{"title":"Dynamic Path Based DNN Synergistic Inference Acceleration in Edge Computing Environment","authors":"Mengpu Zhou, Bowen Zhou, Huitian Wang, Fang Dong, Wei Zhao","doi":"10.1109/ICPADS53394.2021.00076","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00076","url":null,"abstract":"Deep Neural Networks (DNNs) have achieved excellent performance in intelligent applications. Nevertheless, it is elusive for devices with limited resources to support computationally intensive DNNs, while employing the cloud may lead to prohibitive latency. Better solutions are exploiting edge computing and reducing unnecessary computation. Multi-exit DNN based on the early exit mechanism has an impressive effect in the latter, and in edge computing paradigm, model partition on multi-exit chain DNNs is proved to accelerate inference effectively. However, despite reducing computations to some extent, multiple exits may lead to instability of performance due to variable sample quality, performance inferior to the original model especially in the worst case. Furthermore, nowadays DNNs are universally characterized by a directed acyclic graph (DAG), complicating the partition of multi-exit DNN exceedingly. To solve the issues, in this paper, considering online exit prediction and model execution optimization for multi-exit DNN, we propose a Dynamic Path based DNN Synergistic inference acceleration framework (DPDS), where exit designators are designed to avoid iterative entry for exits; to further promote computational synergy in the edge, the multi-exit DNN is dynamically partitioned according to network environment to achieve fine-grained computing offloading. Experimental results show that DPDS can significantly accelerate DNN inference by 1.87× to 6.78×.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133988044","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}

{"title":"Performance Analysis of Open-Source Hypervisors for Automotive Systems","authors":"Zhengjun Zhang, Yanqiang Liu, Jiangtao Chen, Zhengwei Qi, Yifeng Zhang, Huai Liu","doi":"10.1109/ICPADS53394.2021.00072","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00072","url":null,"abstract":"Nowadays, automotive products are intelligence intensive and thus inevitably handle multiple functionalities under the current high-speed networking environment. The embedded virtualization has high potentials in the automotive industry, thanks to its advantages in function integration, resource utilization, and security. The invention of ARM virtualization extensions has made it possible to run open-source hypervisors, such as Xen and KVM, for embedded applications. Nevertheless, there is little work to investigate the performance of these hypervisors on automotive platforms. This paper presents a detailed analysis of different types of open-source hypervisors that can be applied in the ARM platform. We carry out the virtualization performance experiment from the perspectives of CPU, memory, file I/O, and some OS operation performance on Xen and Jailhouse. A series of microbenchmark programs have been designed, specifically to evaluate the real-time performance of various hypervisors and the relevant overhead. Compared with Xen, Jailhouse has better latency performance, stable latency, and little interference jitter. The performance experiment results help us summarize the advantages and disadvantages of these hypervisors in automotive applications.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"207 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131495161","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}

{"title":"Deep Reinforcement Agent for Failure-aware Job scheduling in High-Performance Computing","authors":"K. Yang, Rongyu Cao, Yueyuan Zhou, Jiawei Zhang, En Shao, Guangming Tan","doi":"10.1109/ICPADS53394.2021.00061","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00061","url":null,"abstract":"Job scheduling is crucial in high-performance computing (HPC), which is dedicated to deciding when and which jobs are allocated to the system and placing the jobs on which resources, by considering multiple scheduling goals. Along with the incremental of various resources and dazzling deep learning training (DLT) workloads, job failure becomes a quite common issue in HPC, which will affect user satisfaction and cluster utilization. To alleviate the influence of hardware and software errors as much as possible, in this paper, we aim to tackle the problem of failure-aware job scheduling in HPC clusters. Inspired by the success of previous studies of deep reinforcement learning-driven job scheduling, we propose a novel HPC scheduling agent named FARS (Failure-aware RL-based scheduler) by considering the effects of job failures. On the one hand, a neural network is applied to map the information of raw cluster and job states to job placement decisions. On the other hand, to consider the influence of job failure for user satisfaction and cluster utilization, FARS leverages make-span of the entire workload as the training objective. Additionally, effective exploration and experience replay techniques are applied to obtain effectively converged agent. To evaluate the capability of FARS, we design extensive trace-based simulation experiments with the popular DLT workloads. The experimental results show that, compared with the best baseline model, FARS obtains 5.69% improvement of average make-span under different device error rates. Together, our FARS is an ideal candidate for failure-aware job scheduler in HPC clusters.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"206 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133336121","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}

{"title":"A Forecasting Method of Dual Traffic Condition Indicators Based on Ensemble Learning","authors":"Chuanhao Dong, Zhiqiang Lv, Jianbo Li","doi":"10.1109/ICPADS53394.2021.00047","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00047","url":null,"abstract":"By the prediction of traffic conditions, the occurrence of traffic congestion can be warned in advance, so that the traffic managers can intervene in time, which can help to reduce the risk of traffic congestion. Therefore, aiming at the problem of traffic congestion, a prediction method for dual traffic condition indicators is proposed. The method for capturing spatial dependence based on the topology of roads and road driving direction is proposed to provide more flexible and targeted spatial features for predicting traffic conditions. In addition, according to the real-time and accuracy requirements of traffic conditions prediction, a novel model named dual-channel convolution block is designed to capture the temporal dependence of traffic conditions. Learning from the idea of ensemble learning, $K$ independent base models are trained to predict traffic condition at the same time, and a model fusion mechanism based on real-time traffic conditions is proposed to fuse the predictions of the base models so that the model can have stronger generalization ability to adapt to various noise data in real traffic conditions. The proposed method is validated on the traffic data sets and compares with the optimal model of all the existing models, the proposed method reduces MAPE of speed prediction by 12.1% and TTI prediction by 10.4%.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114355427","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}

{"title":"Trusted Sliding-Window Aggregation over Blockchains","authors":"Qifeng Shao, Zhao Zhang, Cheqing Jin, Aoying Zhou","doi":"10.1109/ICPADS53394.2021.00038","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00038","url":null,"abstract":"Blockchain that continuously generates infinite transactions is widely applied to many decentralized applications. Applications generally focus more on the most recent transaction data to discover trends and make predictions, and thus there is an increasing demand for sliding-window aggregation over blockchains (e.g., a continuous query for the moving average of Bitcoin transaction volume over the last 24 hours). Blockchain submits transactions by block periodically, which makes it work well for sliding-window aggregation. However, the mutual distrust between blockchain nodes makes users consider both query efficiency and query authentication (e.g., simple payment verification (SPV) in Bitcoin). Aggregate B-tree can process sliding-window aggregation in a multi-query setting efficiently. In order to achieve authenticated sliding-window aggregation, a naive scheme may incorporate the Merkle tree into the aggregate B-tree, but that will complicate the index structure, and couple query logic and verification logic. In this paper, we propose a novel authenticated sliding-window aggregation scheme that separates query authentication from query processing. By designing a separate encoded Merkle tree, verification logic can authenticate query results of the aggregate B-tree by itself, without affecting query logic. We also develop an optimized scheme based on FiBA and software guard extensions (SGX), which further reduces aggregate and digest update costs. Security analysis and empirical study validate the robustness and practicality of the proposed scheme.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125738370","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}

{"title":"ShadowDroid: Practical Black-box Attack against ML-based Android Malware Detection","authors":"Jin Zhang, Chennan Zhang, Xiangyu Liu, Yuncheng Wang, Wenrui Diao, Shanqing Guo","doi":"10.1109/ICPADS53394.2021.00084","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00084","url":null,"abstract":"Machine learning (ML) techniques have been widely deployed in the field of Android malware detection. On the other hand, ML-based malware detection also faces the threat of adversarial attacks. Recently, some research has demonstrated the possibility of such attacks under the settings of white-box or grey-box. However, a more practical threat model - black-box adversarial attack has not been well validated and evaluated. In this paper, we bridge this research gap and propose a black-box adversarial attack approach, ShadowDroid, against ML-based Android malware detection. On a high level, ShadowDroid tries to construct a substitute model of the target malware detection system. Utilizing this substitute model, we can identify and modify the key features of a malicious app to generate an adversarial sample. During the experiment, we evaluated the effectiveness of ShadowDroid against nine ML-based Android malware detection frameworks. It achieved successful malware evading on five platforms. Based on these results, we also discuss how to design a robust malware detection system to prevent adversarial attacks.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128693759","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}