Pub Date : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00087
Filip Krakowski, Fabian Ruhland, M. Schöttner
In this paper, we propose Infinileap, a modern networking framework enabling high-performance memory transfer mechanisms like Remote Direct Memory Access (RDMA) for applications written in Java. Infinileap is based on the Open Communication X (UCX) framework, which is accessed from Java. This is accomplished through Oracle's Project Panama, which is currently in the preview phase and aims to significantly improve interoperability between Java and “foreign” languages, such as C. In contrast to often used internal and unsupported JDK APIs, Project Panama's APIs are explicitly intended for use and developers are encouraged to adapt their existing code accordingly. Using Project Panama, we implement an object as well as future-oriented framework based on UCX. Our experiments show that Infinileap and thus Project Panama's innovations work reliably and efficiently under heavy load and also, within benchmarks implemented for this purpose based on the Java Microbenchmark Harness (JMH), achieve very good performance results with over 110 million messages per second and round-trip latencies below two microseconds with a single ConnectX-5 InfiniBand (single-port) network interface controller.
{"title":"Infinileap: Modern High-Performance Networking for Distributed Java Applications based on RDMA","authors":"Filip Krakowski, Fabian Ruhland, M. Schöttner","doi":"10.1109/ICPADS53394.2021.00087","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00087","url":null,"abstract":"In this paper, we propose Infinileap, a modern networking framework enabling high-performance memory transfer mechanisms like Remote Direct Memory Access (RDMA) for applications written in Java. Infinileap is based on the Open Communication X (UCX) framework, which is accessed from Java. This is accomplished through Oracle's Project Panama, which is currently in the preview phase and aims to significantly improve interoperability between Java and “foreign” languages, such as C. In contrast to often used internal and unsupported JDK APIs, Project Panama's APIs are explicitly intended for use and developers are encouraged to adapt their existing code accordingly. Using Project Panama, we implement an object as well as future-oriented framework based on UCX. Our experiments show that Infinileap and thus Project Panama's innovations work reliably and efficiently under heavy load and also, within benchmarks implemented for this purpose based on the Java Microbenchmark Harness (JMH), achieve very good performance results with over 110 million messages per second and round-trip latencies below two microseconds with a single ConnectX-5 InfiniBand (single-port) network interface controller.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"56 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":"124506779","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}
This paper studies the performance bottleneck of tree-based wireless sensor networks. Based on our findings, we propose a collaborative transmission paradigm which opportunistically shifts some node traffics to intermediate links beyond the tree topology. We experimentally demonstrate that the quality of intermediate links can even out over multiple transmissions. Low-Power-Listening based MACs can increase the packet reception ratio of data delivery, but may also introduce asymmetry issues on intermediate link, leading to redundant packet transmissions. To overcome the problem, we select good-SINR links that ensure high reliability with at most $k$ retransmissions for communication. We compute the ratio of tree-link and intermediate long-link transmissions in a distributed way, aiming at minimizing the maximum load in the neighborhood. We implement the method in TinyOS as an independent component named LLC, and evaluate LLC via both simulation and testbed experiments. Results show that LLC can reduce the energy consumption by up to 50%, while retaining the high retransmission reliability.
{"title":"Collaborative Transmission over Intermediate Links in Duty-Cycle WSNs","authors":"Qianwu Chen, Xianjin Xia, Zhigang Li, Yuanqing Zheng","doi":"10.1109/ICPADS53394.2021.00111","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00111","url":null,"abstract":"This paper studies the performance bottleneck of tree-based wireless sensor networks. Based on our findings, we propose a collaborative transmission paradigm which opportunistically shifts some node traffics to intermediate links beyond the tree topology. We experimentally demonstrate that the quality of intermediate links can even out over multiple transmissions. Low-Power-Listening based MACs can increase the packet reception ratio of data delivery, but may also introduce asymmetry issues on intermediate link, leading to redundant packet transmissions. To overcome the problem, we select good-SINR links that ensure high reliability with at most $k$ retransmissions for communication. We compute the ratio of tree-link and intermediate long-link transmissions in a distributed way, aiming at minimizing the maximum load in the neighborhood. We implement the method in TinyOS as an independent component named LLC, and evaluate LLC via both simulation and testbed experiments. Results show that LLC can reduce the energy consumption by up to 50%, while retaining the high retransmission reliability.","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":"114247829","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 : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00106
Leyi Sun, Yifan Zhuo, O. Marin
Scientific applications can benefit greatly from getting deployed on a cloud computing platform, but such deployments require skills and expertise that are beyond the reach of many scientists. We address this issue with a framework that simplifies the process of writing cloud-ready scientific applications, and that automates their deployment and execution on top of cloud infrastructures. This paper presents (1) our domain-specific language whose syntax is simple to learn and use, and (2) our compiler that exploits potential data parallelism opportunities and handles load balancing automatically for the users. Our framework prototype demonstrates the feasibility of our approach, and our scalability analysis looks promising.
{"title":"Simple yet Efficient Deployment of Scientific Applications in the Cloud","authors":"Leyi Sun, Yifan Zhuo, O. Marin","doi":"10.1109/ICPADS53394.2021.00106","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00106","url":null,"abstract":"Scientific applications can benefit greatly from getting deployed on a cloud computing platform, but such deployments require skills and expertise that are beyond the reach of many scientists. We address this issue with a framework that simplifies the process of writing cloud-ready scientific applications, and that automates their deployment and execution on top of cloud infrastructures. This paper presents (1) our domain-specific language whose syntax is simple to learn and use, and (2) our compiler that exploits potential data parallelism opportunities and handles load balancing automatically for the users. Our framework prototype demonstrates the feasibility of our approach, and our scalability analysis looks promising.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"11 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":"114384981","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 : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00012
Yao Xiao, Jiangang Shu, Xiaohua Jia, Hejiao Huang
Federated Learning enables the collaborative learning in cross-client scenarios while keeping the clients' data local for privacy. The presence of non-IID data is one of major challenges in federated learning. To deal with this statistic challenge, federated multi-task learning considers the local training for each client as a single task. However, all the clients must participate in each training round, and it is inapplicable to mobile or IOT devices with constrained communication capability. To achieve the communication-efficiency and high accuracy with non-IID data, we propose a clustered federated multi-task learning by exploring client clustering and multi-task learning. We measure the similarities of local data among clients indirectly through their models' parameters, and design a client clustering strategy to enable clients with similar data distribution into a same group. The limitation of full-participation can be eliminated through the way of model training for groups instead of individual clients. The convergence analysis and experimental evaluation on real-world datasets shows that our work outperforms the basic federated learning in accuracy and is also more communication-efficient than the existing federated multi-task learning.
{"title":"Clustered Federated Multi-Task Learning with Non-IID Data","authors":"Yao Xiao, Jiangang Shu, Xiaohua Jia, Hejiao Huang","doi":"10.1109/ICPADS53394.2021.00012","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00012","url":null,"abstract":"Federated Learning enables the collaborative learning in cross-client scenarios while keeping the clients' data local for privacy. The presence of non-IID data is one of major challenges in federated learning. To deal with this statistic challenge, federated multi-task learning considers the local training for each client as a single task. However, all the clients must participate in each training round, and it is inapplicable to mobile or IOT devices with constrained communication capability. To achieve the communication-efficiency and high accuracy with non-IID data, we propose a clustered federated multi-task learning by exploring client clustering and multi-task learning. We measure the similarities of local data among clients indirectly through their models' parameters, and design a client clustering strategy to enable clients with similar data distribution into a same group. The limitation of full-participation can be eliminated through the way of model training for groups instead of individual clients. The convergence analysis and experimental evaluation on real-world datasets shows that our work outperforms the basic federated learning in accuracy and is also more communication-efficient than the existing federated multi-task learning.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"150 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":"115186135","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 : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00083
Junfeng Zhao, Lu Liu, Feng Liu
Cloud Storage is the fundamental service which is widely used by users of cloud computing. Cloud offers many advantages such as flexibility, elasticity, scalability and data sharing among users. However, the physical separation of cloud storage data and users brings many data security issues. This article focuses on the solution of the security issues, which contains the access control of cloud storage data and the illegal copy of the use process. A watermark embedding and CP-ABE encryption model based on orthogonal operation domain is proposed. Based on this model, a blockchain-based cloud storage data access control and anti-copy scheme is proposed. In order to illustrate the feasibility of the model and the scheme, a watermark embedding and CP-ABE encryption method based on orthogonal operation domain is designed for image type data. Experiments have been carried out to prove the feasibility of the method and scheme.
{"title":"Access Control and Anti-copy Scheme of Cloud Storage Data Based on Blockchain","authors":"Junfeng Zhao, Lu Liu, Feng Liu","doi":"10.1109/ICPADS53394.2021.00083","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00083","url":null,"abstract":"Cloud Storage is the fundamental service which is widely used by users of cloud computing. Cloud offers many advantages such as flexibility, elasticity, scalability and data sharing among users. However, the physical separation of cloud storage data and users brings many data security issues. This article focuses on the solution of the security issues, which contains the access control of cloud storage data and the illegal copy of the use process. A watermark embedding and CP-ABE encryption model based on orthogonal operation domain is proposed. Based on this model, a blockchain-based cloud storage data access control and anti-copy scheme is proposed. In order to illustrate the feasibility of the model and the scheme, a watermark embedding and CP-ABE encryption method based on orthogonal operation domain is designed for image type data. Experiments have been carried out to prove the feasibility of the method and scheme.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"3 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":"115189905","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 : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00006
Lin Li, Lei Wang, Bin Han, Xinxin Lu, Zhiyi Zhou, Bingxian Lu
WiFi-based human activity recognition has been widely used in many fields such as health diagnosis, intrusion detection and smart home. Most existing recognition methods can achieve a satisfying accuracy only in one domain, but low accuracy occurs when models are trained in source domain but are used in target domain. Meanwhile, considering finetuning network directly is impossible or easy to overfit with limited labeled target data, transfer learning based methods with domain adaptive layers are proposed to solve above problems but just aligning marginal distribution, which may lose massive fine-grained features. Based on this, we present an end-to-end deep subdomain adaptive network based activities recognition (DSANAR) using Channel State Information (CSI) that aligns marginal and matches conditional distribution simultaneously for more fine-grained features in each category of relevant subdomains based on a local maximum mean discrepancy (LMMD). Besides, by using a joint cross-entropy and an adaptive loss as training loss, DSANAR outperforms other state-of-art methods on an autonomous dataset with average 95.6% cross-domain accuracy.
{"title":"Subdomain Adaptive Learning Network for Cross-Domain Human Activities Recognition Using WiFi with CSI","authors":"Lin Li, Lei Wang, Bin Han, Xinxin Lu, Zhiyi Zhou, Bingxian Lu","doi":"10.1109/ICPADS53394.2021.00006","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00006","url":null,"abstract":"WiFi-based human activity recognition has been widely used in many fields such as health diagnosis, intrusion detection and smart home. Most existing recognition methods can achieve a satisfying accuracy only in one domain, but low accuracy occurs when models are trained in source domain but are used in target domain. Meanwhile, considering finetuning network directly is impossible or easy to overfit with limited labeled target data, transfer learning based methods with domain adaptive layers are proposed to solve above problems but just aligning marginal distribution, which may lose massive fine-grained features. Based on this, we present an end-to-end deep subdomain adaptive network based activities recognition (DSANAR) using Channel State Information (CSI) that aligns marginal and matches conditional distribution simultaneously for more fine-grained features in each category of relevant subdomains based on a local maximum mean discrepancy (LMMD). Besides, by using a joint cross-entropy and an adaptive loss as training loss, DSANAR outperforms other state-of-art methods on an autonomous dataset with average 95.6% cross-domain accuracy.","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":"122842473","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 : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00088
Tao Li, Xu Cao, Haisong Liu, Chenqi Shi, Pengpeng Chen
As one of the important methods of identity recognition, gait recognition has a wide range of applications in the fields of new human-computer interaction, smart home, smart office and health monitoring. In this paper, we propose a system for multi-person gait recognition (MTPGait) with spatio-temporal information via millimeter wave radar. We specially design a neural network that can extract multi-scale spatio-temporal features along space and time dimensions of 3D point cloud concisely and efficiently. In addition, we construct and release a millimeter wave radar 3D point cloud data set, which consists of 960-minute gait data of 25 volunteers. The experimental results show that MTPGait is able to achieve 96.7% recognition accuracy in a single-person scene on random routes, and 90.2 % recognition accuracy when two people coexist, while the accuracy of the existing methods can not reach 90 % in either scenario.
{"title":"MTPGait: Multi-person Gait Recognition with Spatio-temporal Information via Millimeter Wave Radar","authors":"Tao Li, Xu Cao, Haisong Liu, Chenqi Shi, Pengpeng Chen","doi":"10.1109/ICPADS53394.2021.00088","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00088","url":null,"abstract":"As one of the important methods of identity recognition, gait recognition has a wide range of applications in the fields of new human-computer interaction, smart home, smart office and health monitoring. In this paper, we propose a system for multi-person gait recognition (MTPGait) with spatio-temporal information via millimeter wave radar. We specially design a neural network that can extract multi-scale spatio-temporal features along space and time dimensions of 3D point cloud concisely and efficiently. In addition, we construct and release a millimeter wave radar 3D point cloud data set, which consists of 960-minute gait data of 25 volunteers. The experimental results show that MTPGait is able to achieve 96.7% recognition accuracy in a single-person scene on random routes, and 90.2 % recognition accuracy when two people coexist, while the accuracy of the existing methods can not reach 90 % in either scenario.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"184 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":"124661553","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}
Ethereum is essentially a transaction-driven state machine, and a smart contract is a piece of executable code on Ethereum. Compared with the scripting language on Bitcoin, the smart contract language solidity, which is Turing-complete and the ex-pressive capabilities are very powerful. However, this attribute also brings many potential security threats, vulnerabilities, and various other issues. In this paper, we propose a novel smart contract security technology, named Jyane, to detect the Reentrancy vulnerability, which is one of the most threatening vulnerabilities to smart contracts. More importantly, Our tool-Jyane is the first path profiling solution for smart contracts. Firstly, we use EVM (Ethereum Virtual Machine) binary bytecode to construct control flow graphs (CFG), then use the improved Ball-Larus Path profiling algorithm (BLPP) to generate IDs for acyclic paths. Finally, after profiling the constructed paths, the suspicious paths can be detected successfully. We evaluate Jyane and other technology through comprehensive test and comparison; the results show that Jyane can profile the actual execution path of smart contracts to detect vulnerabilities with a low false-positive rate accurately. From the results of the evaluation, Jyane marked 27 of 1,226 Ethereum smart contracts selected in 2016 and 2017 as vulnerable contracts, included the vulnerability of the DAO contract which once led to a $60 million loss. Furthermore, compared with some other existing detection tools, Jyane shows broader detection range for Reentrancy vulnerabilities with lower time overhead.
{"title":"Jyane: Detecting Reentrancy vulnerabilities based on path profiling method","authors":"Yicheng Fang, Chunping Wang, Zhe Sun, Hongbing Cheng","doi":"10.1109/ICPADS53394.2021.00040","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00040","url":null,"abstract":"Ethereum is essentially a transaction-driven state machine, and a smart contract is a piece of executable code on Ethereum. Compared with the scripting language on Bitcoin, the smart contract language solidity, which is Turing-complete and the ex-pressive capabilities are very powerful. However, this attribute also brings many potential security threats, vulnerabilities, and various other issues. In this paper, we propose a novel smart contract security technology, named Jyane, to detect the Reentrancy vulnerability, which is one of the most threatening vulnerabilities to smart contracts. More importantly, Our tool-Jyane is the first path profiling solution for smart contracts. Firstly, we use EVM (Ethereum Virtual Machine) binary bytecode to construct control flow graphs (CFG), then use the improved Ball-Larus Path profiling algorithm (BLPP) to generate IDs for acyclic paths. Finally, after profiling the constructed paths, the suspicious paths can be detected successfully. We evaluate Jyane and other technology through comprehensive test and comparison; the results show that Jyane can profile the actual execution path of smart contracts to detect vulnerabilities with a low false-positive rate accurately. From the results of the evaluation, Jyane marked 27 of 1,226 Ethereum smart contracts selected in 2016 and 2017 as vulnerable contracts, included the vulnerability of the DAO contract which once led to a $60 million loss. Furthermore, compared with some other existing detection tools, Jyane shows broader detection range for Reentrancy vulnerabilities with lower time overhead.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"5 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":"122399510","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 : 2021-12-01DOI: 10.1109/ICPADS53394.2021.00010
Xinlin Li, Shuqi Liu, Xinyi Zhang, Linqi Song
Traditionally, decision making in the stock market greatly depends on human expertise in sophisticated information processing, such as analyzing various financial reports and related news. However, limitations of expertise, time, and resources make investors suffer from information overload and information imbalance and may pose a negative impact on the investment market. Recent improvements in computing power, the availability of large volumes of data, and the advanced Artificial Intelligence (AI) techniques empower us with the ability to assist decision making in the stock market. In this paper, we present an integrated system that comprehensively monitors the downside risks of individual stocks and the overall market. Specifically, the stock downside risk is predicted based on quantitative data of related stocks, where the relationship between stocks is measured by constructing an Enterprise Knowledge Graph (EKG) using public knowledge. On the other hand, the market downside risk is predicted based on information extracted from daily news. For each risk, a Temporal Convolutional Network (TCN) is trained to output a continuous risk level that reveals both the direction and amplitude of incoming changes. Finally, key information and the predicted risk levels are organized into a condensed and understandable dashboard to interact with investors. Experiments on three focal stocks in the U.S. market suggest convincing accuracy in both stock risk and market risk modeling. Further visualization analysis demonstrates that our model has the potential to inform reverse changes of a stock movement ten days in advance.
{"title":"Predicting Downside in Stock Market Using Knowledge and News Data","authors":"Xinlin Li, Shuqi Liu, Xinyi Zhang, Linqi Song","doi":"10.1109/ICPADS53394.2021.00010","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00010","url":null,"abstract":"Traditionally, decision making in the stock market greatly depends on human expertise in sophisticated information processing, such as analyzing various financial reports and related news. However, limitations of expertise, time, and resources make investors suffer from information overload and information imbalance and may pose a negative impact on the investment market. Recent improvements in computing power, the availability of large volumes of data, and the advanced Artificial Intelligence (AI) techniques empower us with the ability to assist decision making in the stock market. In this paper, we present an integrated system that comprehensively monitors the downside risks of individual stocks and the overall market. Specifically, the stock downside risk is predicted based on quantitative data of related stocks, where the relationship between stocks is measured by constructing an Enterprise Knowledge Graph (EKG) using public knowledge. On the other hand, the market downside risk is predicted based on information extracted from daily news. For each risk, a Temporal Convolutional Network (TCN) is trained to output a continuous risk level that reveals both the direction and amplitude of incoming changes. Finally, key information and the predicted risk levels are organized into a condensed and understandable dashboard to interact with investors. Experiments on three focal stocks in the U.S. market suggest convincing accuracy in both stock risk and market risk modeling. Further visualization analysis demonstrates that our model has the potential to inform reverse changes of a stock movement ten days in advance.","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":"126494228","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}
The basic I/O operations of a system can be categorized as two distinct modes: synchronous (sync) I/O and asynchronous (async) I/O, whose performance varies on the system statues, workloads and storage devices. Appropriately applying I/O modes is critical to the system performance. However, the I/O access of diverse applications in a server, especially in a cloud, is volatile and irregular. As a result, this can lack a flexible and adaptive I/O modes, leading to the sub-optimal I/O performance. To tackle this problem, in this paper, we propose IObrain, an intelligent I/O mode recommendation system, which can adopt the appropriate I/O mode in a dynamic and self-adaptive manner according to both application needs and system statuses. IObrain first trains a lightweight recommendation model with decision tree. Then, a query hook is interposed into the storage engine to intercept the read/write operations from upper application. In this way, IObrain queries the recommendation model first before executing a read/write operation to find the right I/O mode. In addition, two techniques, called inference cache and gRPC bridge, are proposed to reduce the inherent query latency. We practically implement IObrain and verify the advantage of IObrain based on the prototype system. The experimental results show that, compared to existing approach, IObrain improves the I/O performance by up to 1.33× with mild running costs.
{"title":"IObrain: An Intelligent Lightweight I/O Recommendation System based on Decision Tree","authors":"Yiting Huang, Zhiwen Wang, Yuguo Li, Junlang Huang, Dingding Li, Yong Tang, Deze Zeng","doi":"10.1109/ICPADS53394.2021.00011","DOIUrl":"https://doi.org/10.1109/ICPADS53394.2021.00011","url":null,"abstract":"The basic I/O operations of a system can be categorized as two distinct modes: synchronous (sync) I/O and asynchronous (async) I/O, whose performance varies on the system statues, workloads and storage devices. Appropriately applying I/O modes is critical to the system performance. However, the I/O access of diverse applications in a server, especially in a cloud, is volatile and irregular. As a result, this can lack a flexible and adaptive I/O modes, leading to the sub-optimal I/O performance. To tackle this problem, in this paper, we propose IObrain, an intelligent I/O mode recommendation system, which can adopt the appropriate I/O mode in a dynamic and self-adaptive manner according to both application needs and system statuses. IObrain first trains a lightweight recommendation model with decision tree. Then, a query hook is interposed into the storage engine to intercept the read/write operations from upper application. In this way, IObrain queries the recommendation model first before executing a read/write operation to find the right I/O mode. In addition, two techniques, called inference cache and gRPC bridge, are proposed to reduce the inherent query latency. We practically implement IObrain and verify the advantage of IObrain based on the prototype system. The experimental results show that, compared to existing approach, IObrain improves the I/O performance by up to 1.33× with mild running costs.","PeriodicalId":309508,"journal":{"name":"2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS)","volume":"4 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":"125231164","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：