Pub Date : 2021-09-01DOI: 10.1109/EDGE53862.2021.00022
Shangguang Wang, Qing Li, Mengwei Xu, Xiao Ma, Ao Zhou, Qibo Sun
Satellite network is the first step towards interstellar voyages. It can provide global Internet connectivity everywhere on the earth, where most areas cannot access the Internet by the terrestrial infrastructure due to the geographic accessibility and high deployment cost. The space industry experiences a rise in large low-earth-orbit satellite constellations to achieve universal connectivity. The research community is also urgent to do some leading research to bridge the connectivity divide. Researchers now conduct their work by simulation, which is far from enough. However, experiments on real satellites are hindered by the exceptionally high bar of space technology, such as deployment cost and unknown risks. To solve the above challenges, we are eager to contribute to the universal connectivity and build an open research platform, Tiansuan constellation, to support experiments on real satellite networks. We discuss the potential research topics that would benefit from Tiansuan. We provide two case studies that have already been deployed in two experimental satellites of Tiansuan.
{"title":"Tiansuan Constellation: An Open Research Platform","authors":"Shangguang Wang, Qing Li, Mengwei Xu, Xiao Ma, Ao Zhou, Qibo Sun","doi":"10.1109/EDGE53862.2021.00022","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00022","url":null,"abstract":"Satellite network is the first step towards interstellar voyages. It can provide global Internet connectivity everywhere on the earth, where most areas cannot access the Internet by the terrestrial infrastructure due to the geographic accessibility and high deployment cost. The space industry experiences a rise in large low-earth-orbit satellite constellations to achieve universal connectivity. The research community is also urgent to do some leading research to bridge the connectivity divide. Researchers now conduct their work by simulation, which is far from enough. However, experiments on real satellites are hindered by the exceptionally high bar of space technology, such as deployment cost and unknown risks. To solve the above challenges, we are eager to contribute to the universal connectivity and build an open research platform, Tiansuan constellation, to support experiments on real satellite networks. We discuss the potential research topics that would benefit from Tiansuan. We provide two case studies that have already been deployed in two experimental satellites of Tiansuan.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125534932","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00020
M. Ganguli, Sunku Ranganath, Subhiksha Ravisundar, Abhirupa Layek, Dakshina Ilangovan, Edwin Verplanke
As Edge deployments move closer towards the end devices, low latency communication among Edge aware applications is one of the key tenants of Edge service offerings. In order to simplify application development, service mesh architectures have emerged as the evolutionary architectural paradigms for taking care of bulk of application communication logic such as health checks, circuit breaking, secure communication, resiliency (among others), thereby decoupling application logic with communication infrastructure. The latency to throughput ratio needs to be measurable for high performant deployments at the Edge. Providing benchmark data for various edge deployments with Bare Metal and virtual machine-based scenarios, this paper digs into architectural complexities of deploying service mesh at edge environment, performance impact across north-south and east-west communications in and out of a service mesh leveraging popular open-source service mesh Istio/Envoy using a simple on-prem Kubernetes cluster. The performance results shared indicate performance impact of Kubernetes network stack with Envoy data plane. Microarchitecture analyses indicate bottlenecks in Linux based stacks from a CPU micro-architecture perspective and quantify the high impact of Linux's Iptables rule matching at scale. We conclude with the challenges in multiple areas of profiling and benchmarking requirement and a call to action for deploying a service mesh, in latency sensitive environments at Edge.
{"title":"Challenges and Opportunities in Performance Benchmarking of Service Mesh for the Edge","authors":"M. Ganguli, Sunku Ranganath, Subhiksha Ravisundar, Abhirupa Layek, Dakshina Ilangovan, Edwin Verplanke","doi":"10.1109/EDGE53862.2021.00020","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00020","url":null,"abstract":"As Edge deployments move closer towards the end devices, low latency communication among Edge aware applications is one of the key tenants of Edge service offerings. In order to simplify application development, service mesh architectures have emerged as the evolutionary architectural paradigms for taking care of bulk of application communication logic such as health checks, circuit breaking, secure communication, resiliency (among others), thereby decoupling application logic with communication infrastructure. The latency to throughput ratio needs to be measurable for high performant deployments at the Edge. Providing benchmark data for various edge deployments with Bare Metal and virtual machine-based scenarios, this paper digs into architectural complexities of deploying service mesh at edge environment, performance impact across north-south and east-west communications in and out of a service mesh leveraging popular open-source service mesh Istio/Envoy using a simple on-prem Kubernetes cluster. The performance results shared indicate performance impact of Kubernetes network stack with Envoy data plane. Microarchitecture analyses indicate bottlenecks in Linux based stacks from a CPU micro-architecture perspective and quantify the high impact of Linux's Iptables rule matching at scale. We conclude with the challenges in multiple areas of profiling and benchmarking requirement and a call to action for deploying a service mesh, in latency sensitive environments at Edge.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126783909","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00012
Longchuan Yan, Yan Li, Hu Song, Hao Zou, Lijun Wang
With the increase of data center scale, monitoring the status of its internal running devices has become the basis of automatic operations. Traditional environmental monitoring is often based on fixed sensors with limited server racks to collect temperature and humidity data, which cannot be effectively combined with the server task scheduling and resource allocation system. In order to provide reliable services to users, a reliable and low-power data center environmental data monitoring system is needed to collect and analyze data such as temperature, humidity and smoke. This paper presents a data center monitoring system based on edge computing, which uses edge computing and wireless sensor network technology to monitor the running status of the data center. The edge device collects the environmental data in real time, and then obtains the real-time running state parameters of the server through Intelligent Platform Management Interface. By analyzing the environmental data of the data center, the energy consumption and operating parameters of the server can be monitored and adjusted.
{"title":"Edge Computing Based Data Center Monitoring","authors":"Longchuan Yan, Yan Li, Hu Song, Hao Zou, Lijun Wang","doi":"10.1109/EDGE53862.2021.00012","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00012","url":null,"abstract":"With the increase of data center scale, monitoring the status of its internal running devices has become the basis of automatic operations. Traditional environmental monitoring is often based on fixed sensors with limited server racks to collect temperature and humidity data, which cannot be effectively combined with the server task scheduling and resource allocation system. In order to provide reliable services to users, a reliable and low-power data center environmental data monitoring system is needed to collect and analyze data such as temperature, humidity and smoke. This paper presents a data center monitoring system based on edge computing, which uses edge computing and wireless sensor network technology to monitor the running status of the data center. The edge device collects the environmental data in real time, and then obtains the real-time running state parameters of the server through Intelligent Platform Management Interface. By analyzing the environmental data of the data center, the energy consumption and operating parameters of the server can be monitored and adjusted.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133046992","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00024
Juan Fang, Siqi Chen, Min Cai
Edge computing provides more rapid and convenient services to the user by deploying computing resources and storage resources on network edges closer to the user. However, the edge server has small storage capacity, irregular user requests and real-time changes in user preferences. To address these problems, this paper presents a Mobile Edge Data Cooperative Cache Admission Based on Content Popularity (DCCCP) based on the perspective of the content provider. First, we analyze and learn the key feature properties of video objects to build the tree data structure and dynamically adjust the tree structure according to the state of the leaf nodes. Next, the multiarm bandit model is considered for the tree structure characteristics and the number of samples. In addition, considering the limited edge server capacity and the large cloudedge transmission latency, edge collaboration is used for data cache. Finally, we experiment the DCCCP algorithm with four excellent algorithms in terms of hit rate, latency and system cost, and demonstrate the effectiveness of the DCCCP algorithm.
{"title":"Mobile Edge Data Cooperative Cache Admission Based on Content Popularity","authors":"Juan Fang, Siqi Chen, Min Cai","doi":"10.1109/EDGE53862.2021.00024","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00024","url":null,"abstract":"Edge computing provides more rapid and convenient services to the user by deploying computing resources and storage resources on network edges closer to the user. However, the edge server has small storage capacity, irregular user requests and real-time changes in user preferences. To address these problems, this paper presents a Mobile Edge Data Cooperative Cache Admission Based on Content Popularity (DCCCP) based on the perspective of the content provider. First, we analyze and learn the key feature properties of video objects to build the tree data structure and dynamically adjust the tree structure according to the state of the leaf nodes. Next, the multiarm bandit model is considered for the tree structure characteristics and the number of samples. In addition, considering the limited edge server capacity and the large cloudedge transmission latency, edge collaboration is used for data cache. Finally, we experiment the DCCCP algorithm with four excellent algorithms in terms of hit rate, latency and system cost, and demonstrate the effectiveness of the DCCCP algorithm.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121837030","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00011
Handuo Zhang, Jun Na, Bin Zhang
It becomes common to deploy a pre-trained machine learning model on the edge devices to improve their intelligence. Considering the dynamic nature of the edge environment, for ensuring decision accuracy, edge devices always need to collect the latest samples and upload them to the cloud to get an updated model. During this process, it is crucial to determine which samples are necessary to be uploaded considering the communication cost. We propose a scenario adaptive edge data reduction strategy to filter samples differently from existing approaches by measuring whether they can affect current decision accuracy. First, we put forward a novel adaptive data reduction framework for cloud-edge collaborative scenarios. Then, we present the implementation algorithms for filtering samples based on scenarios, identifying candidate scenarios emerging in edge environments, and updating edge scenarios. Experiment results show that in the best case, our approach can discard 70% samples while keeping the same inference accuracy with the original sample set.
{"title":"Scenario Adaptive Edge Data Reduction","authors":"Handuo Zhang, Jun Na, Bin Zhang","doi":"10.1109/EDGE53862.2021.00011","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00011","url":null,"abstract":"It becomes common to deploy a pre-trained machine learning model on the edge devices to improve their intelligence. Considering the dynamic nature of the edge environment, for ensuring decision accuracy, edge devices always need to collect the latest samples and upload them to the cloud to get an updated model. During this process, it is crucial to determine which samples are necessary to be uploaded considering the communication cost. We propose a scenario adaptive edge data reduction strategy to filter samples differently from existing approaches by measuring whether they can affect current decision accuracy. First, we put forward a novel adaptive data reduction framework for cloud-edge collaborative scenarios. Then, we present the implementation algorithms for filtering samples based on scenarios, identifying candidate scenarios emerging in edge environments, and updating edge scenarios. Experiment results show that in the best case, our approach can discard 70% samples while keeping the same inference accuracy with the original sample set.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133395054","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00016
The Bao Bui, Aly Sakr, Juan Castrillón, Rolf Schuster
The growth of Internet of Things (IoT) devices and their unpredictable needs make resource allocation of edge computing systems challenging. A good edge computing system or platform should not only solve the resources allocation challenge to balance loads among edge servers with the best quality of service for all clients but also deal with emergencies where high-priority clients need access to the edge. This paper presents an improvement of an existing algorithm Score-Based Match-Making (SBMM) to solve the aforementioned challenge. A six-factors score-based Match-Making algorithm is proposed to tackle priority-related challenges in resource allocation with a preemption factor to deal with emergency problems. An evaluation of our own orchestration platform (Edge Diagnostics Platform) under different scenarios and algorithms, namely random, naive, SBMM is presented. The experimental studies highlight the improvement in clients' priority distribution in edge servers and solve the problem of emergency clients with preemption. The simulation results verify that the proposed algorithm is significantly better than the original algorithm in the context of prioritized deployments.
{"title":"Six-factors Score-based Match-making Based on Priority and Preemption for Resource Allocation in Edge Computing","authors":"The Bao Bui, Aly Sakr, Juan Castrillón, Rolf Schuster","doi":"10.1109/EDGE53862.2021.00016","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00016","url":null,"abstract":"The growth of Internet of Things (IoT) devices and their unpredictable needs make resource allocation of edge computing systems challenging. A good edge computing system or platform should not only solve the resources allocation challenge to balance loads among edge servers with the best quality of service for all clients but also deal with emergencies where high-priority clients need access to the edge. This paper presents an improvement of an existing algorithm Score-Based Match-Making (SBMM) to solve the aforementioned challenge. A six-factors score-based Match-Making algorithm is proposed to tackle priority-related challenges in resource allocation with a preemption factor to deal with emergency problems. An evaluation of our own orchestration platform (Edge Diagnostics Platform) under different scenarios and algorithms, namely random, naive, SBMM is presented. The experimental studies highlight the improvement in clients' priority distribution in edge servers and solve the problem of emergency clients with preemption. The simulation results verify that the proposed algorithm is significantly better than the original algorithm in the context of prioritized deployments.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127475435","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00019
Sihai Tang, B. Chen, Harold Iwen, Jason Hirsch, Song Fu, Qing Yang, P. Palacharla, N. Wang, Xi Wang, Weisong Shi
Autonomous vehicle systems require sensor data to make crucial driving and traffic management decisions. Reliable data as well as computational resources become critical. In this paper, we develop a Vehicular Edge Computing FRAMEwork (VECFrame) for connected and autonomous vehicles (CAVs) exploring containerization, indirect communication, and edge-enabled cooperative object detection. Through our framework, the data, generated by on-board sensors, is used towards various edge serviceable tasks. Due to the limited view of a vehicle, sensor data from one vehicle cannot be used to perceive road and traffic condition of a larger area. To address this problem, VECFrame facilitates data transfer and fusion and cooperative object detection from multiple vehicles. Through real-world experiments, we evaluate the performance and robustness of our framework on different device architectures and under different scenarios. We demonstrate that our framework achieves a more accurate perception of traffic condition via vehicle-edge data transfer and on-edge computation.
{"title":"VECFrame: A Vehicular Edge Computing Framework for Connected Autonomous Vehicles","authors":"Sihai Tang, B. Chen, Harold Iwen, Jason Hirsch, Song Fu, Qing Yang, P. Palacharla, N. Wang, Xi Wang, Weisong Shi","doi":"10.1109/EDGE53862.2021.00019","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00019","url":null,"abstract":"Autonomous vehicle systems require sensor data to make crucial driving and traffic management decisions. Reliable data as well as computational resources become critical. In this paper, we develop a Vehicular Edge Computing FRAMEwork (VECFrame) for connected and autonomous vehicles (CAVs) exploring containerization, indirect communication, and edge-enabled cooperative object detection. Through our framework, the data, generated by on-board sensors, is used towards various edge serviceable tasks. Due to the limited view of a vehicle, sensor data from one vehicle cannot be used to perceive road and traffic condition of a larger area. To address this problem, VECFrame facilitates data transfer and fusion and cooperative object detection from multiple vehicles. Through real-world experiments, we evaluate the performance and robustness of our framework on different device architectures and under different scenarios. We demonstrate that our framework achieves a more accurate perception of traffic condition via vehicle-edge data transfer and on-edge computation.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125995288","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 : 2021-09-01DOI: 10.1109/EDGE53862.2021.00017
Mohamed Abdulmaksoud, Ninad Dehadrai, Juan Castrillón, Aly Sakr, Rolf Schuster
The increasing demand for low-latency high-performance applications motivates the development of network and compute infrastructure. As an emerging paradigm, edge computing is becoming the chosen solution for many low-latency applications in many industries. However, the current orches-tration and diagnostics methods do not fulfill the requirements of the new edge computing architectures. In contrast to cloud computing, edge applications are very sensitive to changes in the infrastructure. And thus, the orchestration and diagnosis of the infrastructure must be aware of the edge application's special needs. In this research work, we present a solution model: The Edge Diagnostics Platform. The platform has two main functions: Orchestration and Diagnosis. We show the design principles of the platform, how it can help with the orchestration and diagnosis of edge applications. Finally, we carry out practical experiments to show how the platform may be used to diagnose network and CPU problems. The results show practically accurate detection of network and CPU problems.
{"title":"Edge Diagnostics Platform: Orchestration and Diagnosis Model for Edge Computing Infrastructure","authors":"Mohamed Abdulmaksoud, Ninad Dehadrai, Juan Castrillón, Aly Sakr, Rolf Schuster","doi":"10.1109/EDGE53862.2021.00017","DOIUrl":"https://doi.org/10.1109/EDGE53862.2021.00017","url":null,"abstract":"The increasing demand for low-latency high-performance applications motivates the development of network and compute infrastructure. As an emerging paradigm, edge computing is becoming the chosen solution for many low-latency applications in many industries. However, the current orches-tration and diagnostics methods do not fulfill the requirements of the new edge computing architectures. In contrast to cloud computing, edge applications are very sensitive to changes in the infrastructure. And thus, the orchestration and diagnosis of the infrastructure must be aware of the edge application's special needs. In this research work, we present a solution model: The Edge Diagnostics Platform. The platform has two main functions: Orchestration and Diagnosis. We show the design principles of the platform, how it can help with the orchestration and diagnosis of edge applications. Finally, we carry out practical experiments to show how the platform may be used to diagnose network and CPU problems. The results show practically accurate detection of network and CPU problems.","PeriodicalId":115969,"journal":{"name":"2021 IEEE International Conference on Edge Computing (EDGE)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130187087","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：