Pub Date : 2024-03-20DOI: 10.1109/TSUSC.2024.3379550
Duc Van Le;Jing Zhou;Rongrong Wang;Rui Tan;Fei Duan
Data centers (DCs) are power-intensive facilities which use a significant amount of energy for cooling the servers. Increasing the temperature and relative humidity (RH) setpoints is a rule-of-thumb approach to reducing the DC energy usage. However, the high temperature and RH may undermine the server's reliability. Before we can choose the proper temperature and RH settings, it is essential to understand how the temperature and RH setpoints affect the DC power usage and server's reliability. To this end, we constructed and experimented with an air-cooled DC testbed in Singapore, which consists of a direct expansion cooling system and 521 servers running real-world application workloads. This paper presents the key measurement results and observations from our 11-month experiments. Our results suggest that by operating at a supply air temperature setpoints of 29 $^{circ }$ C, our testbed achieves substantial cooling power saving with little impact on the server's reliability. Furthermore, we present a total cost of ownership (TCO) analysis framework which guides settings of the temperature and RH for a DC. Our observations and TCO analysis framework will be useful to future efforts in building and operating air-cooled DCs in tropics and beyond.
数据中心（DC）是电力密集型设施，需要消耗大量能源来冷却服务器。提高温度和相对湿度（RH）设定值是减少 DC 能源消耗的一个常用方法。然而，高温和相对湿度可能会降低服务器的可靠性。在选择合适的温度和相对湿度设置之前，我们必须了解温度和相对湿度设置点如何影响直流电能使用和服务器的可靠性。为此，我们在新加坡建造了一个风冷直流试验台并进行了实验，该试验台由直接膨胀冷却系统和运行实际应用工作负载的 521 台服务器组成。本文介绍了为期 11 个月实验的主要测量结果和观察结果。我们的结果表明，通过在 29$^{circ }$C 的供气温度设定值下运行，我们的测试平台实现了大量的制冷节能，而对服务器的可靠性影响很小。此外，我们还提出了一个总拥有成本（TCO）分析框架，用于指导直流电的温度和相对湿度设置。我们的观察结果和总拥有成本分析框架将有助于今后在热带地区及其他地区建造和运行风冷直流电。
{"title":"Impacts of Increasing Temperature and Relative Humidity in Air-Cooled Tropical Data Centers","authors":"Duc Van Le;Jing Zhou;Rongrong Wang;Rui Tan;Fei Duan","doi":"10.1109/TSUSC.2024.3379550","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3379550","url":null,"abstract":"Data centers (DCs) are power-intensive facilities which use a significant amount of energy for cooling the servers. Increasing the temperature and relative humidity (RH) setpoints is a rule-of-thumb approach to reducing the DC energy usage. However, the high temperature and RH may undermine the server's reliability. Before we can choose the proper temperature and RH settings, it is essential to understand how the temperature and RH setpoints affect the DC power usage and server's reliability. To this end, we constructed and experimented with an air-cooled DC testbed in Singapore, which consists of a direct expansion cooling system and 521 servers running real-world application workloads. This paper presents the key measurement results and observations from our 11-month experiments. Our results suggest that by operating at a supply air temperature setpoints of 29\u0000<inline-formula><tex-math>$^{circ }$</tex-math></inline-formula>\u0000C, our testbed achieves substantial cooling power saving with little impact on the server's reliability. Furthermore, we present a total cost of ownership (TCO) analysis framework which guides settings of the temperature and RH for a DC. Our observations and TCO analysis framework will be useful to future efforts in building and operating air-cooled DCs in tropics and beyond.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 5","pages":"790-802"},"PeriodicalIF":3.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-17DOI: 10.1109/TSUSC.2024.3401003
Kangshun Li;Shumin Xie;Tianjin Zhu;Hui Wang
With the development of technology, unmanned aerial vehicles (UAVs) and Internet of Things devices are widely used in smart agriculture, resulting in significant energy consumption. In this paper, the optimization problem for UAV-assisted mobile computing in smart agriculture is modeled as a constrained multiobjective optimization problem. By jointly optimizing the deployment position of UAVs, the offloading location of the tasks, the transmit power of the devices, and the resource allocation of the UAVs, two optimization objectives (total delay and energy consumption) are minimized simultaneously. In view of the complex constraints, a constrained multiobjective algorithm named JO-DPTS is proposed. The algorithm adopts dual-population and two-stage approach to improve population convergence and diversity. The simulation results substantiate that JO-DPTS exhibits superior performance compared to the other three state-of-the-art constrained multiobjective evolutionary algorithms.
{"title":"Constrained Multiobjective Optimization for UAV-Assisted Mobile Edge Computing in Smart Agriculture: Minimizing Delay and Energy Consumption","authors":"Kangshun Li;Shumin Xie;Tianjin Zhu;Hui Wang","doi":"10.1109/TSUSC.2024.3401003","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3401003","url":null,"abstract":"With the development of technology, unmanned aerial vehicles (UAVs) and Internet of Things devices are widely used in smart agriculture, resulting in significant energy consumption. In this paper, the optimization problem for UAV-assisted mobile computing in smart agriculture is modeled as a constrained multiobjective optimization problem. By jointly optimizing the deployment position of UAVs, the offloading location of the tasks, the transmit power of the devices, and the resource allocation of the UAVs, two optimization objectives (total delay and energy consumption) are minimized simultaneously. In view of the complex constraints, a constrained multiobjective algorithm named JO-DPTS is proposed. The algorithm adopts dual-population and two-stage approach to improve population convergence and diversity. The simulation results substantiate that JO-DPTS exhibits superior performance compared to the other three state-of-the-art constrained multiobjective evolutionary algorithms.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 6","pages":"948-957"},"PeriodicalIF":3.0,"publicationDate":"2024-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810513","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Data privacy breaches have prompted growing concerns regarding privacy issues on social networks. Preserving the privacy of links in the directed social graph, where edges signify the information flow or data contributions, poses a formidable challenge. However, existing methods for uncertain graphs primarily target undirected graphs and lack rigorous privacy guarantees. In this paper, we present a personal evidence protection algorithm called PEPA, which provides formally dual privacy guarantees for directed social links. Specifically, we implement out-link privacy to protect the out-links of nodes. Despite this protection, the exposure of in-links can still compromise privacy, potentially affecting service quality. To address this, we further introduce an uncertain directed graph algorithm as a post-processing approach for out-link privacy. This algorithm injects uncertainty into nodes’ in-links, effectively transforming the original directed graph into a probability-driven uncertain structure. Additionally, we propose an effective noise optimization method. Finally, we evaluate the trade-off between privacy and utility achieved by PEPA through comparative experiments. The results demonstrate privacy enhancements of PEPA compared to the $(k, varepsilon )$-obfuscation algorithm and utility improvements over the RandWalk algorithm and UG-NDP. Particularly, PEPA demonstrates approximately a 2-fold improvement in utility compared to PEPA without noise optimization.
{"title":"Preserving Link Privacy in Uncertain Directed Social Graphs With Formal Guarantees","authors":"Jiajun Chen;Chunqiang Hu;Ruifeng Zhao;Shaojiang Deng;Xiaoshuang Xing;Jiguo Yu","doi":"10.1109/TSUSC.2024.3399754","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3399754","url":null,"abstract":"Data privacy breaches have prompted growing concerns regarding privacy issues on social networks. Preserving the privacy of links in the directed social graph, where edges signify the information flow or data contributions, poses a formidable challenge. However, existing methods for uncertain graphs primarily target undirected graphs and lack rigorous privacy guarantees. In this paper, we present a personal evidence protection algorithm called PEPA, which provides formally dual privacy guarantees for directed social links. Specifically, we implement out-link privacy to protect the out-links of nodes. Despite this protection, the exposure of in-links can still compromise privacy, potentially affecting service quality. To address this, we further introduce an uncertain directed graph algorithm as a post-processing approach for out-link privacy. This algorithm injects uncertainty into nodes’ in-links, effectively transforming the original directed graph into a probability-driven uncertain structure. Additionally, we propose an effective noise optimization method. Finally, we evaluate the trade-off between privacy and utility achieved by PEPA through comparative experiments. The results demonstrate privacy enhancements of PEPA compared to the <inline-formula><tex-math>$(k, varepsilon )$</tex-math></inline-formula>-obfuscation algorithm and utility improvements over the RandWalk algorithm and UG-NDP. Particularly, PEPA demonstrates approximately a 2-fold improvement in utility compared to PEPA without noise optimization.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"10 1","pages":"108-119"},"PeriodicalIF":3.0,"publicationDate":"2024-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184033","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-08DOI: 10.1109/TSUSC.2024.3371544
Ke Yuan;Peng Sang;Jian Ge;Bingcai Zhou;Chunfu Jia
E-Voting is widely used in many social, economic, political and cultural fields for its convenience, efficiency and greenness, but how to guarantee the fairness of e-voting and the controllability of human intervention needs further in-depth research and exploration. Although the introduction of homomorphic encryption algorithm solves the problem of ballot privacy calculation, and most of these schemes solve the problem of private key confidentiality by using or overlaying multiple different methods of saving private keys, its security will be questioned as long as there is a possibility of human intervention in the saving process. To solve this problem, we propose a timed-release e-voting scheme based on Paillier homomorphic encryption. We analyze the semantic security of the ballot formally by defining the security game, and realize the legitimacy check of the ballot ciphertext through the idea of partial knowledge proof. Property analysis shows that this scheme satisfies the basic properties of the security requirements of the e-voting scheme. Performance analysis shows that this scheme is feasible to implement in practical voting.
{"title":"A Timed-Release E-Voting Scheme Based on Paillier Homomorphic Encryption","authors":"Ke Yuan;Peng Sang;Jian Ge;Bingcai Zhou;Chunfu Jia","doi":"10.1109/TSUSC.2024.3371544","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3371544","url":null,"abstract":"E-Voting is widely used in many social, economic, political and cultural fields for its convenience, efficiency and greenness, but how to guarantee the fairness of e-voting and the controllability of human intervention needs further in-depth research and exploration. Although the introduction of homomorphic encryption algorithm solves the problem of ballot privacy calculation, and most of these schemes solve the problem of private key confidentiality by using or overlaying multiple different methods of saving private keys, its security will be questioned as long as there is a possibility of human intervention in the saving process. To solve this problem, we propose a timed-release e-voting scheme based on Paillier homomorphic encryption. We analyze the semantic security of the ballot formally by defining the security game, and realize the legitimacy check of the ballot ciphertext through the idea of partial knowledge proof. Property analysis shows that this scheme satisfies the basic properties of the security requirements of the e-voting scheme. Performance analysis shows that this scheme is feasible to implement in practical voting.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 5","pages":"740-753"},"PeriodicalIF":3.0,"publicationDate":"2024-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-06DOI: 10.1109/TSUSC.2024.3374049
Jinbo Wang;Ruijin Wang;Guangquan Xu;Donglin He;Xikai Pei;Fengli Zhang;Jie Gan
Federated learning is a distributed learning paradigm, which is usually combined with edge computing to meet the joint training of IoT devices. A significant challenge in federated learning lies in the statistical heterogeneity, characterized by non-independent and identically distributed (non-IID) local data across diverse parties. This heterogeneity can result in inconsistent optimization within individual local models. Although previous research has endeavored to tackle issues stemming from heterogeneous data, our findings indicate that these attempts have not yielded high-performance neural network models. To overcome this fundamental challenge, we introduce the framework called FedPKR in this paper, which facilitates efficient federated learning through knowledge review. The core principle of FedPKR involves leveraging the knowledge representation generated by the global and local model layers to conduct periodic layer-by-layer comparative learning in a reciprocal manner. This strategy rectifies local model training, leading to enhanced outcomes. Our experimental results and subsequent analysis substantiate that FedPKR effectively augments model accuracy in image classification tasks, meanwhile demonstrating resilience to statistical heterogeneity across all participating entities.
{"title":"FedPKR: Federated Learning With Non-IID Data via Periodic Knowledge Review in Edge Computing","authors":"Jinbo Wang;Ruijin Wang;Guangquan Xu;Donglin He;Xikai Pei;Fengli Zhang;Jie Gan","doi":"10.1109/TSUSC.2024.3374049","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3374049","url":null,"abstract":"Federated learning is a distributed learning paradigm, which is usually combined with edge computing to meet the joint training of IoT devices. A significant challenge in federated learning lies in the statistical heterogeneity, characterized by non-independent and identically distributed (non-IID) local data across diverse parties. This heterogeneity can result in inconsistent optimization within individual local models. Although previous research has endeavored to tackle issues stemming from heterogeneous data, our findings indicate that these attempts have not yielded high-performance neural network models. To overcome this fundamental challenge, we introduce the framework called FedPKR in this paper, which facilitates efficient federated learning through knowledge review. The core principle of FedPKR involves leveraging the knowledge representation generated by the global and local model layers to conduct periodic layer-by-layer comparative learning in a reciprocal manner. This strategy rectifies local model training, leading to enhanced outcomes. Our experimental results and subsequent analysis substantiate that FedPKR effectively augments model accuracy in image classification tasks, meanwhile demonstrating resilience to statistical heterogeneity across all participating entities.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 6","pages":"902-912"},"PeriodicalIF":3.0,"publicationDate":"2024-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-03-02DOI: 10.1109/TSUSC.2024.3396381
Soteris Constantinou;Constantinos Costa;Andreas Konstantinidis;Panos K. Chrysanthis;Demetrios Zeinalipour-Yazti
The escalating global energy crisis and the increasing ${text{CO}_{2}}$ emissions have necessitated the optimization of energy efficiency. The proliferation of Internet of Things (IoTs) devices, expected to reach 100 billion by 2030, contributed to this energy crisis and subsequently to the global ${text{CO}_{2}}$ emissions increase. Concomitantly, climate and energy targets have paved the way for an escalating adoption of solar photovoltaic power generation in residences. The IoT integration into home energy management systems holds the potential to yield energy and peak demand savings. Optimizing device planning to mitigate ${text{CO}_{2}}$ emissions poses significant challenges due to the complexity of user-defined preferences and consumption patterns. In this article, we propose an innovative IoT data platform, coined Sustainable Energy Management Framework (SEMF), which aims to balance the trade-off between the imported energy from the grid, users’ comfort, and ${text{CO}_{2}}$ emissions. SEMF incorporates a Green Planning evolutionary algorithm, coined GreenCap$^+$+, to facilitate load shifting of IoT-enabled devices, taking into consideration the integration of renewable energy sources, multiple constraints, peak-demand times, and dynamic pricing. Based on our experimental evaluation utilizing real-world data, our prototype system has outperformed the state-of-the-art approach by up to $approx$29% reduction in imported energy, $approx$35% increase in self-consumption of renewable energy, and $approx$34% decrease in ${text{CO}_{2}}$ emissions, while maintaining a high level of user comfort $approx$94%-99%.
{"title":"A Sustainable Energy Management Framework for Smart Homes","authors":"Soteris Constantinou;Constantinos Costa;Andreas Konstantinidis;Panos K. Chrysanthis;Demetrios Zeinalipour-Yazti","doi":"10.1109/TSUSC.2024.3396381","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3396381","url":null,"abstract":"The escalating global energy crisis and the increasing <inline-formula><tex-math>${text{CO}_{2}}$</tex-math></inline-formula> emissions have necessitated the optimization of energy efficiency. The proliferation of Internet of Things (IoTs) devices, expected to reach 100 billion by 2030, contributed to this energy crisis and subsequently to the global <inline-formula><tex-math>${text{CO}_{2}}$</tex-math></inline-formula> emissions increase. Concomitantly, climate and energy targets have paved the way for an escalating adoption of solar photovoltaic power generation in residences. The IoT integration into home energy management systems holds the potential to yield energy and peak demand savings. Optimizing device planning to mitigate <inline-formula><tex-math>${text{CO}_{2}}$</tex-math></inline-formula> emissions poses significant challenges due to the complexity of user-defined preferences and consumption patterns. In this article, we propose an innovative IoT data platform, coined <i>Sustainable Energy Management Framework (SEMF)</i>, which aims to balance the trade-off between the imported energy from the grid, users’ comfort, and <inline-formula><tex-math>${text{CO}_{2}}$</tex-math></inline-formula> emissions. <i>SEMF</i> incorporates a Green Planning evolutionary algorithm, coined <i>GreenCap<inline-formula><tex-math>$^+$</tex-math><alternatives><mml:math><mml:msup><mml:mrow/><mml:mo>+</mml:mo></mml:msup></mml:math><inline-graphic></alternatives></inline-formula></i>, to facilitate load shifting of IoT-enabled devices, taking into consideration the integration of renewable energy sources, multiple constraints, peak-demand times, and dynamic pricing. Based on our experimental evaluation utilizing real-world data, our prototype system has outperformed the state-of-the-art approach by up to <inline-formula><tex-math>$approx$</tex-math></inline-formula>29% reduction in imported energy, <inline-formula><tex-math>$approx$</tex-math></inline-formula>35% increase in self-consumption of renewable energy, and <inline-formula><tex-math>$approx$</tex-math></inline-formula>34% decrease in <inline-formula><tex-math>${text{CO}_{2}}$</tex-math></inline-formula> emissions, while maintaining a high level of user comfort <inline-formula><tex-math>$approx$</tex-math></inline-formula>94%-99%.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"10 1","pages":"70-81"},"PeriodicalIF":3.0,"publicationDate":"2024-03-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143184031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-28DOI: 10.1109/TSUSC.2024.3370837
Vidushi Agarwal;Omid Ardakanian;Sujata Pal
With the rollout of smart meters, a vast amount of energy time-series became available from homes, enabling applications such as non-intrusive load monitoring (NILM). The inconspicuous collection of this data, however, poses a risk to the privacy of customers. Federated Learning (FL) eliminates the problem of sharing raw data with a cloud service provider by allowing machine learning models to be trained in a collaborative fashion on decentralized data. Although several NILM techniques that rely on FL to train a deep neural network for identifying the energy consumption of individual appliances have been proposed in recent years, the robustness of these techniques to malicious users and their ability to fully protect the user privacy remain unexplored. In this paper, we present a robust and privacy-preserving FL-based framework to train a bidirectional transformer architecture for NILM. This framework takes advantage of a meta-learning algorithm to handle the data heterogeneity prevalent in real-world settings. The efficacy of the proposed framework is corroborated through comparative experiments using two real-world NILM datasets. The results show that this framework can attain an accuracy that is on par with a centrally-trained energy disaggregation model, while preserving user privacy.
{"title":"A Robust and Privacy-Aware Federated Learning Framework for Non-Intrusive Load Monitoring","authors":"Vidushi Agarwal;Omid Ardakanian;Sujata Pal","doi":"10.1109/TSUSC.2024.3370837","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3370837","url":null,"abstract":"With the rollout of smart meters, a vast amount of energy time-series became available from homes, enabling applications such as non-intrusive load monitoring (NILM). The inconspicuous collection of this data, however, poses a risk to the privacy of customers. Federated Learning (FL) eliminates the problem of sharing raw data with a cloud service provider by allowing machine learning models to be trained in a collaborative fashion on decentralized data. Although several NILM techniques that rely on FL to train a deep neural network for identifying the energy consumption of individual appliances have been proposed in recent years, the robustness of these techniques to malicious users and their ability to fully protect the user privacy remain unexplored. In this paper, we present a robust and privacy-preserving FL-based framework to train a bidirectional transformer architecture for NILM. This framework takes advantage of a meta-learning algorithm to handle the data heterogeneity prevalent in real-world settings. The efficacy of the proposed framework is corroborated through comparative experiments using two real-world NILM datasets. The results show that this framework can attain an accuracy that is on par with a centrally-trained energy disaggregation model, while preserving user privacy.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 5","pages":"766-777"},"PeriodicalIF":3.0,"publicationDate":"2024-02-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-23DOI: 10.1109/TSUSC.2024.3369333
Pierre Jacquet;Thomas Ledoux;Romain Rouvoy
Despite continuous improvements, cloud physical resources remain underused, hence severely impacting the efficiency of these infrastructures at large. To overcome this inefficiency, Infrastructure-as-a-Service (IaaS) providers usually compensate for oversized Virtual Machines (VMs) by offering more virtual resources than are physically available on a host. However, this technique—known as oversubscription —may hinder performances when a statically-defined oversubscription ratio results in resource contention of hosted VMs. Therefore, instead of setting a static and cluster-wide ratio, this article studies how a greedy increase of the oversubscription ratio per Physical Machine (PM) and resources type can preserve performance goals. Keeping performance unchanged allows our contribution to be more realistically adopted by production-scale IaaS infrastructures. This contribution, named ScroogeVM , leverages the detection of PM stability to carefully increase the associated oversubscription ratios. Based on metrics shared by public cloud providers, we investigate the impact of resource oversubscription on performance degradation. Subsequently, we conduct a comparative analysis of ScroogeVM with state-of-the-art oversubscription computations. The results demonstrate that our approach outperforms existing methods by leveraging the presence of long-lasting VMs, while avoiding live migration penalties and performance impacts for stakeholders.
{"title":"SCROOGEVM: Boosting Cloud Resource Utilization With Dynamic Oversubscription","authors":"Pierre Jacquet;Thomas Ledoux;Romain Rouvoy","doi":"10.1109/TSUSC.2024.3369333","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3369333","url":null,"abstract":"Despite continuous improvements, cloud physical resources remain underused, hence severely impacting the efficiency of these infrastructures at large. To overcome this inefficiency, Infrastructure-as-a-Service (IaaS) providers usually compensate for oversized Virtual Machines (VMs) by offering more virtual resources than are physically available on a host. However, this technique—known as \u0000<italic>oversubscription</i>\u0000—may hinder performances when a statically-defined oversubscription ratio results in resource contention of hosted VMs. Therefore, instead of setting a static and cluster-wide ratio, this article studies how a greedy increase of the oversubscription ratio per Physical Machine (PM) and resources type can preserve performance goals. Keeping performance unchanged allows our contribution to be more realistically adopted by production-scale IaaS infrastructures. This contribution, named \u0000<sc>ScroogeVM</small>\u0000, leverages the detection of PM stability to carefully increase the associated oversubscription ratios. Based on metrics shared by public cloud providers, we investigate the impact of resource oversubscription on performance degradation. Subsequently, we conduct a comparative analysis of \u0000<sc>ScroogeVM</small>\u0000 with state-of-the-art oversubscription computations. The results demonstrate that our approach outperforms existing methods by leveraging the presence of long-lasting VMs, while avoiding live migration penalties and performance impacts for stakeholders.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 5","pages":"754-765"},"PeriodicalIF":3.0,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-23DOI: 10.1109/TSUSC.2024.3369092
Shuai Guo;Menglei Xia;Huanqun Xue;Shuang Wang;Chao Liu
With the in-depth study of the internal process mechanism of the global ocean by oceanographers, traditional ocean observation methods have been unable to meet the new observation requirements. In order to achieve a low-cost ocean observation mechanism with high spatio-temporal resolution, this paper introduces mobile crowd sensing technology into the field of ocean observation. First, a Transformer-based vessel trajectory prediction algorithm is proposed, which can monitor the location and movement trajectory of vessel in real time. Second, the participant selection algorithm in mobile crowd sensing is studied, and based on the trajectory prediction algorithm, a dynamic participant selection algorithm for ocean mobile crowd sensing is proposed by combining it with the discrete particle swarm optimization (DPSO) algorithm. Third, a coverage estimation algorithm is designed to estimate the coverage of the selection scheme. Finally, the spatio-temporal resolution of the vessel's driving trajectory is analyzed through experiments, which verifies the effectiveness of the algorithm and comprehensively confirms the feasibility of mobile crowd sensing in the field of ocean observation.
{"title":"OceanCrowd: Vessel Trajectory Data-Based Participant Selection for Mobile Crowd Sensing in Ocean Observation","authors":"Shuai Guo;Menglei Xia;Huanqun Xue;Shuang Wang;Chao Liu","doi":"10.1109/TSUSC.2024.3369092","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3369092","url":null,"abstract":"With the in-depth study of the internal process mechanism of the global ocean by oceanographers, traditional ocean observation methods have been unable to meet the new observation requirements. In order to achieve a low-cost ocean observation mechanism with high spatio-temporal resolution, this paper introduces mobile crowd sensing technology into the field of ocean observation. First, a Transformer-based vessel trajectory prediction algorithm is proposed, which can monitor the location and movement trajectory of vessel in real time. Second, the participant selection algorithm in mobile crowd sensing is studied, and based on the trajectory prediction algorithm, a dynamic participant selection algorithm for ocean mobile crowd sensing is proposed by combining it with the discrete particle swarm optimization (DPSO) algorithm. Third, a coverage estimation algorithm is designed to estimate the coverage of the selection scheme. Finally, the spatio-temporal resolution of the vessel's driving trajectory is analyzed through experiments, which verifies the effectiveness of the algorithm and comprehensively confirms the feasibility of mobile crowd sensing in the field of ocean observation.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 6","pages":"889-901"},"PeriodicalIF":3.0,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142810537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-16DOI: 10.1109/TSUSC.2024.3366502
Syed Muhammad Danish;Kaiwen Zhang;Fatima Amara;Juan Carlos Oviedo Cepeda;Luis Fernando Rueda Vasquez;Tom Marynowski
Climate change is a major issue that has disastrous impacts on the environment through different causes like the greenhouse gas (GHG) emission. Many energy utilities around the world intend to reduce GHG emissions by promoting different systems including carbon emission trading (CET), renewable energy certificates (RECs), and tradable white certificates (TWCs). However, these systems are centralized, highly regulated, and operationally expensive and do not meet transparency, trust and security requirements. Accordingly, GHG emission reduction schemes are gradually moving towards blockchain-based solutions due to their underpinning characteristics including decentralization, transparency, anonymity, and trust (independent from third parties). This paper performs a comprehensive investigation into the blockchain technology, deployed for GHG emission reduction plans. It explores existing blockchain solutions along with their associated challenges to effectively uncover their potentials. As a result, this study suggests possible lines of research for future enhancements of blockchain systems particularly their incorporation in GHG emission reduction.
{"title":"Blockchain for Energy Credits and Certificates: A Comprehensive Review","authors":"Syed Muhammad Danish;Kaiwen Zhang;Fatima Amara;Juan Carlos Oviedo Cepeda;Luis Fernando Rueda Vasquez;Tom Marynowski","doi":"10.1109/TSUSC.2024.3366502","DOIUrl":"https://doi.org/10.1109/TSUSC.2024.3366502","url":null,"abstract":"Climate change is a major issue that has disastrous impacts on the environment through different causes like the greenhouse gas (GHG) emission. Many energy utilities around the world intend to reduce GHG emissions by promoting different systems including carbon emission trading (CET), renewable energy certificates (RECs), and tradable white certificates (TWCs). However, these systems are centralized, highly regulated, and operationally expensive and do not meet transparency, trust and security requirements. Accordingly, GHG emission reduction schemes are gradually moving towards blockchain-based solutions due to their underpinning characteristics including decentralization, transparency, anonymity, and trust (independent from third parties). This paper performs a comprehensive investigation into the blockchain technology, deployed for GHG emission reduction plans. It explores existing blockchain solutions along with their associated challenges to effectively uncover their potentials. As a result, this study suggests possible lines of research for future enhancements of blockchain systems particularly their incorporation in GHG emission reduction.","PeriodicalId":13268,"journal":{"name":"IEEE Transactions on Sustainable Computing","volume":"9 5","pages":"727-739"},"PeriodicalIF":3.0,"publicationDate":"2024-02-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142397220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","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：