Sangdae Kim, Cheonyong Kim, Hyunchong Cho, Yongbin Yim, Sang-Ha Kim
The real-time data dissemination is based on a minimum delivery speed of each hop on the end-to-end shortest distance in order to ensure to reach the destination within the desired time deadline in hop-by-hop fashion. In other words, the schemes depend not only on single hop-delay, but also on the distance between the source and destination. However, the performance of the real-time data dissemination could suffered from some "Voids" in the irregular sensor network which blocks the data forwarding. Namely, the voids lengthen the delivery path and the time for detour the voids. These extended delivery path and time aggravate real-time data dissemination success ratio because the additional distance and time increment were not considered in the calculated minimum delivery speed at source node. To deal with this phenomenon, there are detouring methods of voids such as perimeter routing, back pressure. Although the methods enable a void detouring of data packets, the real-time data dissemination success ratio have not improved much because of a many control message to detour voids. To solve this problem, we propose void avoidance scheme for real-time data dissemination. In our scheme, nodes surrounding a void could detect. This void information are transmitted to whole sensors in the network by broadcasting. When a source transmits data to sink, the source uses the void information to find certain point for avoiding voids by the shortest distance. Thus, the source could set desired speed toward the destination via the point without facing a voids. Performance evaluation shows that our scheme provides better real-time dissemination success ratio in practical environments.
{"title":"Void Avoidance Scheme for Real-Time Data Dissemination in Irregular Wireless Sensor Networks","authors":"Sangdae Kim, Cheonyong Kim, Hyunchong Cho, Yongbin Yim, Sang-Ha Kim","doi":"10.1109/AINA.2016.59","DOIUrl":"https://doi.org/10.1109/AINA.2016.59","url":null,"abstract":"The real-time data dissemination is based on a minimum delivery speed of each hop on the end-to-end shortest distance in order to ensure to reach the destination within the desired time deadline in hop-by-hop fashion. In other words, the schemes depend not only on single hop-delay, but also on the distance between the source and destination. However, the performance of the real-time data dissemination could suffered from some \"Voids\" in the irregular sensor network which blocks the data forwarding. Namely, the voids lengthen the delivery path and the time for detour the voids. These extended delivery path and time aggravate real-time data dissemination success ratio because the additional distance and time increment were not considered in the calculated minimum delivery speed at source node. To deal with this phenomenon, there are detouring methods of voids such as perimeter routing, back pressure. Although the methods enable a void detouring of data packets, the real-time data dissemination success ratio have not improved much because of a many control message to detour voids. To solve this problem, we propose void avoidance scheme for real-time data dissemination. In our scheme, nodes surrounding a void could detect. This void information are transmitted to whole sensors in the network by broadcasting. When a source transmits data to sink, the source uses the void information to find certain point for avoiding voids by the shortest distance. Thus, the source could set desired speed toward the destination via the point without facing a voids. Performance evaluation shows that our scheme provides better real-time dissemination success ratio in practical environments.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"56 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122245284","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}
In the previous researches, a great number of routing protocols have been proposed in mobile ad-hoc networks (MANET). But it's hard to be adopted efficiently in low rate-wireless personal area network (LR-WPAN) because there are additional particular constraints of the packet size, bandwidth and so on. To improve the performance in this environment, we propose a Cluster-head and Border-node based Cluster Routing Protocol (CBCRP). CBCRP focuses on the environment that the intra-cluster is stable without frequent topology change. For the intra-cluster routing, we optimize the DSDV which is the proactive way to reduce the end-to-end delay maintaining the low cost of routing overhead. The intra-cluster route information is maintained in both of the cluster head and member nodes. For inter-cluster routing, we optimize the AODV which is the reactive way to reduce the flooding overhead maintaining the network performance. The source cluster-head, destination cluster-head and all the intermediate border nodes are in charge of the inter-cluster route discovery, while only border nodes are in charge of the inter-cluster data transmission. In the simulation result using NS-2, CBCRP shows better performance than the cluster based routing protocol (CBRP), about 21% lower end-to-end delay, 67% lower packet overhead and 60% lower packet loss rate at best case.
{"title":"Cluster-Head and Border-Node Based Cluster Routing Protocol for LR-WPAN","authors":"Ying Huang, Hongli Ge, Jangsu Lee, Yuehua Dai, Dong Xu, Jin Zhang, Q. Gao, Shaoying Zhao","doi":"10.1109/AINA.2016.31","DOIUrl":"https://doi.org/10.1109/AINA.2016.31","url":null,"abstract":"In the previous researches, a great number of routing protocols have been proposed in mobile ad-hoc networks (MANET). But it's hard to be adopted efficiently in low rate-wireless personal area network (LR-WPAN) because there are additional particular constraints of the packet size, bandwidth and so on. To improve the performance in this environment, we propose a Cluster-head and Border-node based Cluster Routing Protocol (CBCRP). CBCRP focuses on the environment that the intra-cluster is stable without frequent topology change. For the intra-cluster routing, we optimize the DSDV which is the proactive way to reduce the end-to-end delay maintaining the low cost of routing overhead. The intra-cluster route information is maintained in both of the cluster head and member nodes. For inter-cluster routing, we optimize the AODV which is the reactive way to reduce the flooding overhead maintaining the network performance. The source cluster-head, destination cluster-head and all the intermediate border nodes are in charge of the inter-cluster route discovery, while only border nodes are in charge of the inter-cluster data transmission. In the simulation result using NS-2, CBCRP shows better performance than the cluster based routing protocol (CBRP), about 21% lower end-to-end delay, 67% lower packet overhead and 60% lower packet loss rate at best case.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"146 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116616339","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}
Mission planning plays an important role in satellite control systems, especially with increase of number of satellites and more complex missions to be planned. In a general setting, the satellite mission scheduling consists in allocating tasks such as observation, communication, etc. to resources (spacecrafts (SCs), satellites, ground stations). For instance, in ground station scheduling the aim is to compute an optimal planning of communications between satellites and operations teams of Ground Station (GS). Because the communication between SCs and GSs can be done during specific window times, this problem can also be seen as a window time scheduling problem. The required communication time is usually quite smaller than the window of visibility of SCs to GSs, however, clashes are produced, making the problem highly constrained. In this work we present a Web interface for solving satellite scheduling problems through various heuristic methods. The Web interface enables the users to remotely solve their problem instances through a selection of heuristic methods such as local search methods (Hill Climbing, Simulated Annealing and Tabu Search) and population-based methods (Genetic Algorithms and variants). The user can select to solve previously generated instances by the STK simulation toolkit or generate their own problem instances. The heuristic methods are easily configurable so that users can simulate a variety of scenarios, problem sizes, etc. The execution of the heuristics methods is done at a HPC Cluster infrastructure supporting efficient execution of various solvers. Additionally, the Web application allows users to keep track of their executions as well as to share problem instances with other users.
{"title":"A Web Interface for Satellite Scheduling Problems","authors":"F. Xhafa, C. García, Admir Barolli, M. Takizawa","doi":"10.1109/AINA.2016.21","DOIUrl":"https://doi.org/10.1109/AINA.2016.21","url":null,"abstract":"Mission planning plays an important role in satellite control systems, especially with increase of number of satellites and more complex missions to be planned. In a general setting, the satellite mission scheduling consists in allocating tasks such as observation, communication, etc. to resources (spacecrafts (SCs), satellites, ground stations). For instance, in ground station scheduling the aim is to compute an optimal planning of communications between satellites and operations teams of Ground Station (GS). Because the communication between SCs and GSs can be done during specific window times, this problem can also be seen as a window time scheduling problem. The required communication time is usually quite smaller than the window of visibility of SCs to GSs, however, clashes are produced, making the problem highly constrained. In this work we present a Web interface for solving satellite scheduling problems through various heuristic methods. The Web interface enables the users to remotely solve their problem instances through a selection of heuristic methods such as local search methods (Hill Climbing, Simulated Annealing and Tabu Search) and population-based methods (Genetic Algorithms and variants). The user can select to solve previously generated instances by the STK simulation toolkit or generate their own problem instances. The heuristic methods are easily configurable so that users can simulate a variety of scenarios, problem sizes, etc. The execution of the heuristics methods is done at a HPC Cluster infrastructure supporting efficient execution of various solvers. Additionally, the Web application allows users to keep track of their executions as well as to share problem instances with other users.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116625854","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}
Everton R. Lira, Enrique Fynn, P. S. Coelho, L. Faina, Lásaro J. Camargos, R. Villaça, R. Pasquini
This paper introduces and evaluates a Traffic Sign Management Architecture (TSMA), which represents a paradigm shift for the deployment of traffic sign infrastructure in the context of Intelligent Transport Systems, Vehicular Networks and Smart Cities. The proposal addresses limitations of the current traffic control model by enabling remote updates of traffic signs and displaying them on the vehicular navigation system display to improve their legibility. TSMA is an architecture developed to provide V2I interaction using a commodity technology, Wi-Fi, through the beacon-stuffing technique. The initial design of TSMA's security mechanisms is also presented in this paper. Evaluations were performed on a developed prototype and simulation environments.
{"title":"An Architecture for Traffic Sign Management in Smart Cities","authors":"Everton R. Lira, Enrique Fynn, P. S. Coelho, L. Faina, Lásaro J. Camargos, R. Villaça, R. Pasquini","doi":"10.1109/AINA.2016.40","DOIUrl":"https://doi.org/10.1109/AINA.2016.40","url":null,"abstract":"This paper introduces and evaluates a Traffic Sign Management Architecture (TSMA), which represents a paradigm shift for the deployment of traffic sign infrastructure in the context of Intelligent Transport Systems, Vehicular Networks and Smart Cities. The proposal addresses limitations of the current traffic control model by enabling remote updates of traffic signs and displaying them on the vehicular navigation system display to improve their legibility. TSMA is an architecture developed to provide V2I interaction using a commodity technology, Wi-Fi, through the beacon-stuffing technique. The initial design of TSMA's security mechanisms is also presented in this paper. Evaluations were performed on a developed prototype and simulation environments.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"120 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121962429","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}
Studying networking technologies has become one of the most important subjects in the field of Technology of junior high schools and Information Study of senior high schools in Japan. To study these subjects, studying with practically and experimentally is strongly recommended by MEXT (Ministry of Education, Culture, Sports, Science and Technologies in Japan). However, there is little equipment for these studies. We have been developing new educational equipment for experimental studying of networking technologies based on concept of physical visualizations and physical direct manipulations. The equipment can visualize physically how network works using LED. Also the equipment provide physical direct manipulation using dials and buttons.
{"title":"New Educational Equipments for Networking Study by Physical Visualizations and Physical Direct Manipurations","authors":"Kazuaki Yoshihara, Kenzi Watanabe, N. Iguchi","doi":"10.1109/AINA.2016.91","DOIUrl":"https://doi.org/10.1109/AINA.2016.91","url":null,"abstract":"Studying networking technologies has become one of the most important subjects in the field of Technology of junior high schools and Information Study of senior high schools in Japan. To study these subjects, studying with practically and experimentally is strongly recommended by MEXT (Ministry of Education, Culture, Sports, Science and Technologies in Japan). However, there is little equipment for these studies. We have been developing new educational equipment for experimental studying of networking technologies based on concept of physical visualizations and physical direct manipulations. The equipment can visualize physically how network works using LED. Also the equipment provide physical direct manipulation using dials and buttons.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130765727","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}
Michael Rethfeldt, P. Danielis, Benjamin Beichler, Björn Konieczek, Felix Uster, D. Timmermann
The new standard IEEE 802.11s enables vendor-independent wireless mesh networks based on the 802.11 WLAN technology. Transmission Control Protocol (TCP) is the most widespread transport protocol for reliable data delivery and still the basis for many network applications. TCP supports different mechanisms for flow and congestion control. However, designed for wired networks, it does not consider the dynamics of wireless networks and especially multi-hop wireless mesh networks. In addition, 802.11s provides own mechanisms such as Automatic Repeat Request (ARQ) for frame retransmissions to hide wireless loss from the upper layers. Being transparent to each other, retransmission schemes on both layers may interfere and operate redundantly, if not properly adjusted. We study the effects of ARQ retry limit variation on TCP throughput in a real-world multi-hop 802.11s test bed. As a result, we suggest ARQ adaptation based on the 802.11s standard's Airtime Link Metric (ALM) for path selection, serving as indicator for overall frame travel time. Our proposed approach solely relies on standard features and imposes no modifications to 802.11s or TCP.
新标准IEEE 802.11s使基于802.11 WLAN技术的独立于供应商的无线网状网络成为可能。传输控制协议(TCP)是用于可靠数据传输的最广泛的传输协议，并且仍然是许多网络应用的基础。TCP支持不同的流量和拥塞控制机制。然而，它是为有线网络设计的，没有考虑无线网络特别是多跳无线网状网络的动态特性。此外，802.11s还提供了自己的机制，例如用于帧重传的自动重复请求(ARQ)，以对上层隐藏无线丢失。由于两层的重传方案相互透明，如果不适当调整，可能会产生干扰和冗余操作。在实际的多跳802.11s测试平台上，研究了ARQ重试限制变化对TCP吞吐量的影响。因此，我们建议基于802.11s标准的Airtime Link Metric (ALM)进行路径选择的ARQ适应，作为整体帧传播时间的指标。我们提出的方法完全依赖于标准特性，不需要对802.11s或TCP进行任何修改。
{"title":"Evaluating Cross-Layer Cooperation of Congestion and Flow Control in IEEE 802.11s Networks","authors":"Michael Rethfeldt, P. Danielis, Benjamin Beichler, Björn Konieczek, Felix Uster, D. Timmermann","doi":"10.1109/AINA.2016.12","DOIUrl":"https://doi.org/10.1109/AINA.2016.12","url":null,"abstract":"The new standard IEEE 802.11s enables vendor-independent wireless mesh networks based on the 802.11 WLAN technology. Transmission Control Protocol (TCP) is the most widespread transport protocol for reliable data delivery and still the basis for many network applications. TCP supports different mechanisms for flow and congestion control. However, designed for wired networks, it does not consider the dynamics of wireless networks and especially multi-hop wireless mesh networks. In addition, 802.11s provides own mechanisms such as Automatic Repeat Request (ARQ) for frame retransmissions to hide wireless loss from the upper layers. Being transparent to each other, retransmission schemes on both layers may interfere and operate redundantly, if not properly adjusted. We study the effects of ARQ retry limit variation on TCP throughput in a real-world multi-hop 802.11s test bed. As a result, we suggest ARQ adaptation based on the 802.11s standard's Airtime Link Metric (ALM) for path selection, serving as indicator for overall frame travel time. Our proposed approach solely relies on standard features and imposes no modifications to 802.11s or TCP.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130090598","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}
Hiroki Kataoka, A. Sawada, Dilawaer Duolikun, T. Enokido, M. Takizawa
It is critical to reduce the electric energy consumed in information systems, especially server clusters. In this paper, we extend the multi-level power consumption (MLPC) model and the multi-level computation (MLC) model to a server with multiple CPUs. In this paper, we newly propose a totally energy-aware (TEA) algorithm to select a server for a process in a cluster. Here, servers in a cluster are first classified into subclusters. Each subcluster is characterized in terms of the electric power and computation rate. One server is randomly selected in each subcluster. Then, one server is selected so that the expected electric energy is minimum in the selected servers. We evaluate the TEA algorithm and show not only the total electric energy consumption of the servers but also the average execution time of processes are reduced in the TEA algorithm compared with other algorithms.
{"title":"Energy-Aware Server Selection Algorithms in a Scalable Cluster","authors":"Hiroki Kataoka, A. Sawada, Dilawaer Duolikun, T. Enokido, M. Takizawa","doi":"10.1109/AINA.2016.154","DOIUrl":"https://doi.org/10.1109/AINA.2016.154","url":null,"abstract":"It is critical to reduce the electric energy consumed in information systems, especially server clusters. In this paper, we extend the multi-level power consumption (MLPC) model and the multi-level computation (MLC) model to a server with multiple CPUs. In this paper, we newly propose a totally energy-aware (TEA) algorithm to select a server for a process in a cluster. Here, servers in a cluster are first classified into subclusters. Each subcluster is characterized in terms of the electric power and computation rate. One server is randomly selected in each subcluster. Then, one server is selected so that the expected electric energy is minimum in the selected servers. We evaluate the TEA algorithm and show not only the total electric energy consumption of the servers but also the average execution time of processes are reduced in the TEA algorithm compared with other algorithms.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127375694","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}
Fault localization is a core element in SDN network management. Many SDN fault reasoning and verification techniques assist operators focus on either analyzing the control plane configuration or checking the data plane network behavior. These solutions are limited in that they cannot correlate network symptoms between the control and the data planes, and are harder to generalize across protocols since they have to model complex configuration languages and dynamic protocol behavior. This paper proposes a new approach called Service Oriented Verification Integrated Reasoning (SOVIR) to tackle SDN fault reasoning. In the SOVIR system, a network user can request one or multiple network services via a high level Service Provisioning Language (SPL). SOVIR automatically parses each provisioned service and presents it as a logical Service View, which consists of a pair of logical end nodes, a service specification, and a list of required network functions (e.g., load balancer). After provisioned in an SDN network, SOVIR queries the controller about the network topology and flow rules from all SDN switches. Based on the flow rules and the configuration of end nodes and network function nodes, SOVIR maps the Service View to an Implementation View, in which all the logical components in the Service View are mapped to the actual system components along with the actual network topology. SOVIR uses an extended Symptom-Fault-Verification model to incorporate various verification techniques systematically into fault reasoning process to localize the faults in SDN. SOVIR has been evaluated in a simulation environment for its accuracy and efficiency. The evaluation shows that with SOVIR, both performance and accuracy of fault reasoning in the simulated SDN networks can be greatly improved by taking properly selected verification tools on specific network entities.
{"title":"Service Oriented Verification Integrated Fault Reasoning for SDNs","authors":"Yongning Tang, Guang Cheng, Zhiwei Xu, F. Chen","doi":"10.1109/AINA.2016.54","DOIUrl":"https://doi.org/10.1109/AINA.2016.54","url":null,"abstract":"Fault localization is a core element in SDN network management. Many SDN fault reasoning and verification techniques assist operators focus on either analyzing the control plane configuration or checking the data plane network behavior. These solutions are limited in that they cannot correlate network symptoms between the control and the data planes, and are harder to generalize across protocols since they have to model complex configuration languages and dynamic protocol behavior. This paper proposes a new approach called Service Oriented Verification Integrated Reasoning (SOVIR) to tackle SDN fault reasoning. In the SOVIR system, a network user can request one or multiple network services via a high level Service Provisioning Language (SPL). SOVIR automatically parses each provisioned service and presents it as a logical Service View, which consists of a pair of logical end nodes, a service specification, and a list of required network functions (e.g., load balancer). After provisioned in an SDN network, SOVIR queries the controller about the network topology and flow rules from all SDN switches. Based on the flow rules and the configuration of end nodes and network function nodes, SOVIR maps the Service View to an Implementation View, in which all the logical components in the Service View are mapped to the actual system components along with the actual network topology. SOVIR uses an extended Symptom-Fault-Verification model to incorporate various verification techniques systematically into fault reasoning process to localize the faults in SDN. SOVIR has been evaluated in a simulation environment for its accuracy and efficiency. The evaluation shows that with SOVIR, both performance and accuracy of fault reasoning in the simulated SDN networks can be greatly improved by taking properly selected verification tools on specific network entities.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":" 15","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114053269","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}
Ouyang Xue, P. Garraghan, D. McKee, P. Townend, Jie Xu
Cloud computing systems face the substantial challenge of the Long Tail problem: a small subset of straggling tasks significantly impede parallel jobs completion. This behavior results in longer service response times and degraded system utilization. Speculative execution, which create task replicas at runtime, is a typical method deployed in large-scale distributed systems to tolerate stragglers. This approach defines stragglers by specifying a static threshold value, which calculates the temporal difference between an individual task and the average task progression for a job. However, specifying static threshold debilitates speculation effectiveness as it fails to consider the intrinsic diversity of job timing constraints within modern day Cloud computing systems. Capturing such heterogeneity enables the ability to impose different levels of strictness for replica creation while achieving specified levels of QoS for different application types. Furthermore, a static threshold also fails to consider system environmental constraints in terms of replication overheads and optimal system resource usage. In this paper we present an algorithm for dynamically calculating a threshold value to identify task stragglers, considering key parameters including job QoS timing constraints, task execution characteristics, and optimal system resource utilization. We study and demonstrate the effectiveness of our algorithm through simulating a number of different operational scenarios based on real production cluster data against state-of-the-art solutions. Results demonstrate that our approach is capable of creating 58.62% less replicas under high resource utilization while reducing response time up to 17.86% for idle periods compared to a static threshold.
{"title":"Straggler Detection in Parallel Computing Systems through Dynamic Threshold Calculation","authors":"Ouyang Xue, P. Garraghan, D. McKee, P. Townend, Jie Xu","doi":"10.1109/AINA.2016.84","DOIUrl":"https://doi.org/10.1109/AINA.2016.84","url":null,"abstract":"Cloud computing systems face the substantial challenge of the Long Tail problem: a small subset of straggling tasks significantly impede parallel jobs completion. This behavior results in longer service response times and degraded system utilization. Speculative execution, which create task replicas at runtime, is a typical method deployed in large-scale distributed systems to tolerate stragglers. This approach defines stragglers by specifying a static threshold value, which calculates the temporal difference between an individual task and the average task progression for a job. However, specifying static threshold debilitates speculation effectiveness as it fails to consider the intrinsic diversity of job timing constraints within modern day Cloud computing systems. Capturing such heterogeneity enables the ability to impose different levels of strictness for replica creation while achieving specified levels of QoS for different application types. Furthermore, a static threshold also fails to consider system environmental constraints in terms of replication overheads and optimal system resource usage. In this paper we present an algorithm for dynamically calculating a threshold value to identify task stragglers, considering key parameters including job QoS timing constraints, task execution characteristics, and optimal system resource utilization. We study and demonstrate the effectiveness of our algorithm through simulating a number of different operational scenarios based on real production cluster data against state-of-the-art solutions. Results demonstrate that our approach is capable of creating 58.62% less replicas under high resource utilization while reducing response time up to 17.86% for idle periods compared to a static threshold.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129276197","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 is concerned with positional estimation of obstacle and aircraft by using ISDB-T signal delays. Recently, Multi-Static Primary Surveillance Radar (MSPSR) has been expected to be used as primary surveillance radar alternative. MSPSR classified into passive bistatic radar uses not only conventional radar signal but also other communication signals. Among them, Digital Terrestrial Television Broadcasting (DTTB) is the most expected radio waves because it has high transmitted power. We have started the feasibility study of MSPSR using DTTB signal. Our system concept is to use original characteristics of DTTB. Since Japanese DTTB employs OFDM and guard interval, delayed signals caused by obstacles can be separated. In this paper, we discuss the positional estimation of obstacles and aircraft. This leads to the validation of the proposed method. Firstly, Japanese DTTB signal types of ISDB-T is introduced. Then, the principle of positional estimation is discussed. Positional estimation is started from detection of delayed signals, and positions are determined by bistatic ranging. Finally, delay profiles and bistatic ranging are shown in experimental results. Those results indicate that our proposed system concept would be useful for estimating obstacle and aircraft positions.
{"title":"Positional Estimation of Obstacles and Aircraft by Using ISDB-T Signal Delay","authors":"J. Honda, T. Otsuyama","doi":"10.1109/AINA.2016.149","DOIUrl":"https://doi.org/10.1109/AINA.2016.149","url":null,"abstract":"This paper is concerned with positional estimation of obstacle and aircraft by using ISDB-T signal delays. Recently, Multi-Static Primary Surveillance Radar (MSPSR) has been expected to be used as primary surveillance radar alternative. MSPSR classified into passive bistatic radar uses not only conventional radar signal but also other communication signals. Among them, Digital Terrestrial Television Broadcasting (DTTB) is the most expected radio waves because it has high transmitted power. We have started the feasibility study of MSPSR using DTTB signal. Our system concept is to use original characteristics of DTTB. Since Japanese DTTB employs OFDM and guard interval, delayed signals caused by obstacles can be separated. In this paper, we discuss the positional estimation of obstacles and aircraft. This leads to the validation of the proposed method. Firstly, Japanese DTTB signal types of ISDB-T is introduced. Then, the principle of positional estimation is discussed. Positional estimation is started from detection of delayed signals, and positions are determined by bistatic ranging. Finally, delay profiles and bistatic ranging are shown in experimental results. Those results indicate that our proposed system concept would be useful for estimating obstacle and aircraft positions.","PeriodicalId":438655,"journal":{"name":"2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116808386","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：