Pub Date : 2014-12-01DOI: 10.1109/PADSW.2014.7097859
A. Jouy, Jianguo Yao, G. Zhu
Avionic networks exerting the Avionics Full-Duplex Switched Ethernet (AFDX) protocol utilize a small amount of the bandwidth to transmit critical traffics. As there is an increasing demand on data exchange for non critical applications, it is of great interest to make use of the physically available capability of the network through optimal bandwidth allocation. In this paper, the problem of bandwidth allocation in AFDX networks is treated in the framework of Network Utility Maximization (NUM). In the present work, multi-path routing is used for non-critical applications to explore the available bandwidth and to improve system performance. The optimization problem is decomposed into a rate update subproblem and a traffic routing subproblem linked together by a pricing dynamic system. A distributed algorithm for bandwidth allocation with multi-path routing is developed and the convergence of the algorithm is proven using Lyapunov stability theory. Some issues related to the implementation of the devolved algorithm in the context of real AFDX networks are addressed and the corresponding solutions are provided. Finally, TrueTime based simulations conform the viability and the applicability of the proposed approach.
{"title":"Optimal bandwidth allocation with dynamic multi-path routing for non-critical traffic in AFDX networks","authors":"A. Jouy, Jianguo Yao, G. Zhu","doi":"10.1109/PADSW.2014.7097859","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097859","url":null,"abstract":"Avionic networks exerting the Avionics Full-Duplex Switched Ethernet (AFDX) protocol utilize a small amount of the bandwidth to transmit critical traffics. As there is an increasing demand on data exchange for non critical applications, it is of great interest to make use of the physically available capability of the network through optimal bandwidth allocation. In this paper, the problem of bandwidth allocation in AFDX networks is treated in the framework of Network Utility Maximization (NUM). In the present work, multi-path routing is used for non-critical applications to explore the available bandwidth and to improve system performance. The optimization problem is decomposed into a rate update subproblem and a traffic routing subproblem linked together by a pricing dynamic system. A distributed algorithm for bandwidth allocation with multi-path routing is developed and the convergence of the algorithm is proven using Lyapunov stability theory. Some issues related to the implementation of the devolved algorithm in the context of real AFDX networks are addressed and the corresponding solutions are provided. Finally, TrueTime based simulations conform the viability and the applicability of the proposed approach.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"85 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114090412","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097857
Huey-Ing Liu, Wen He, Winston K.G. Seah
Ambient energy from solar, vibration, heat and wind provide alternative energy sources to power sensors and extend the lifetime of wireless sensor networks which have traditionally been powered by batteries. This paper aims to enhance the performance of energy harvesting powered wireless sensor networks in three aspects: relaying, scheduling, and medium access control. To better adapt to the characteristics of energy harvesting, an asynchronous receiver-initiated duty-cycling approach is preferred in energy harvesting powered wireless sensor networks. This reduces the duty cycle of senders, and regulates the active and sleep intervals according to the energy levels of sensors. When nodes run out of power and need time to recharge, network holes or voids develop, forcing data packets to be routed via other paths, like detours. The proposed relaying strategy aims to prevent holes by balancing the load across the network according to nodes' energy harvesting characteristics. This is a natural consequence of the asynchronous duty cycling by scheduling transmission based on the receiver's availability. The simulation results show that our scheme outperforms in terms of sender duty cycle, end-to-end delay and delivery ratio, especially in challenged conditions where other protocols fail.
{"title":"LEB-MAC: Load and energy balancing MAC protocol for energy harvesting powered wireless sensor networks","authors":"Huey-Ing Liu, Wen He, Winston K.G. Seah","doi":"10.1109/PADSW.2014.7097857","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097857","url":null,"abstract":"Ambient energy from solar, vibration, heat and wind provide alternative energy sources to power sensors and extend the lifetime of wireless sensor networks which have traditionally been powered by batteries. This paper aims to enhance the performance of energy harvesting powered wireless sensor networks in three aspects: relaying, scheduling, and medium access control. To better adapt to the characteristics of energy harvesting, an asynchronous receiver-initiated duty-cycling approach is preferred in energy harvesting powered wireless sensor networks. This reduces the duty cycle of senders, and regulates the active and sleep intervals according to the energy levels of sensors. When nodes run out of power and need time to recharge, network holes or voids develop, forcing data packets to be routed via other paths, like detours. The proposed relaying strategy aims to prevent holes by balancing the load across the network according to nodes' energy harvesting characteristics. This is a natural consequence of the asynchronous duty cycling by scheduling transmission based on the receiver's availability. The simulation results show that our scheme outperforms in terms of sender duty cycle, end-to-end delay and delivery ratio, especially in challenged conditions where other protocols fail.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125615791","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097877
Yuhua Lin, Haiying Shen
Question and Answer (Q&A) systems aggregate the collected intelligence of all users to provide satisfying answers for questions. A well-developed Q&A system should incorporate features such as high question response rate, high answer quality, a spam-free environment for users. Previous works use reputation systems to achieve the goals. However, these reputation systems evaluate a user with an overall rating for all questions the user has answered regardless of the question categories, thus the reputation score does not accurately reflect the user's ability to answer a question in a specific category. We propose SmartQ: a reputation based Q&A System. SmartQ employs a category and theme based reputation management system to evaluate users' willingness and capability to answer various kinds of questions. The reputation system facilitates the forwarding of a question to favorable experts, which improves the question response rate and answer quality. Also, SmartQ incorporates a lightweight spammer detection method to identify potential spammers. Our trace-driven simulation on PeerSim demonstrates the effectiveness of SmartQ in providing a good user experience. We then develop a real application of SmartQ and deploy it for use in a student group in Clemson University. The user feedback shows that SmartQ can provide high-quality answers for users in a community.
{"title":"SmartQ: A question and answer system for supplying high-quality and trustworthy answers","authors":"Yuhua Lin, Haiying Shen","doi":"10.1109/PADSW.2014.7097877","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097877","url":null,"abstract":"Question and Answer (Q&A) systems aggregate the collected intelligence of all users to provide satisfying answers for questions. A well-developed Q&A system should incorporate features such as high question response rate, high answer quality, a spam-free environment for users. Previous works use reputation systems to achieve the goals. However, these reputation systems evaluate a user with an overall rating for all questions the user has answered regardless of the question categories, thus the reputation score does not accurately reflect the user's ability to answer a question in a specific category. We propose SmartQ: a reputation based Q&A System. SmartQ employs a category and theme based reputation management system to evaluate users' willingness and capability to answer various kinds of questions. The reputation system facilitates the forwarding of a question to favorable experts, which improves the question response rate and answer quality. Also, SmartQ incorporates a lightweight spammer detection method to identify potential spammers. Our trace-driven simulation on PeerSim demonstrates the effectiveness of SmartQ in providing a good user experience. We then develop a real application of SmartQ and deploy it for use in a student group in Clemson University. The user feedback shows that SmartQ can provide high-quality answers for users in a community.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128353853","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097845
Bo Wang, Xinghui Zhao, David Chiu
Limited battery power has long been a challenge for mobile applications. As a result, the work in power monitoring and management has attracted great interests. In this paper, we propose a model to estimate power consumption of mobile applications at run-time, based on application-specific per-action power profiling. In addition, we have developed on-line optimization techniques which help maximize users' experience while conserving power. Our power model is lightweight and flexible, in that it can be used by any mobile applications as a plugin, and it can support user-defined optimization mechanisms. This approach has been evaluated using a case study, a mobile application for field studies, and the experimental results show that our model accurately captures power consumption of the application, and the model can be used to optimize the power consumption based on users' needs.
{"title":"Lightweight online power monitoring and control for mobile applications","authors":"Bo Wang, Xinghui Zhao, David Chiu","doi":"10.1109/PADSW.2014.7097845","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097845","url":null,"abstract":"Limited battery power has long been a challenge for mobile applications. As a result, the work in power monitoring and management has attracted great interests. In this paper, we propose a model to estimate power consumption of mobile applications at run-time, based on application-specific per-action power profiling. In addition, we have developed on-line optimization techniques which help maximize users' experience while conserving power. Our power model is lightweight and flexible, in that it can be used by any mobile applications as a plugin, and it can support user-defined optimization mechanisms. This approach has been evaluated using a case study, a mobile application for field studies, and the experimental results show that our model accurately captures power consumption of the application, and the model can be used to optimize the power consumption based on users' needs.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"72 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130342523","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097906
Jia-Jiun Yang, Jehn-Ruey Jiang, Yung-Liang Lai
This paper studies a Disk Covering Tour Problem (DCTP) for reducing the energy consumption of a mobile robot's movement to provide services for sensor nodes in a wireless sensor network (WSN). Given a set of locations of sensor nodes and a starting location of mobile robot, the DCTP is to find a minimum cost tour of a sequence of tour stops for the mobile robot to serve sensor nodes by keeping every sensor node within a specified distance of a tour stop. We propose an algorithm, called Decreasing k-means (Dk-means), to find an approximate solution to the DCTP. The idea is to select a minimum number of disks or circles of a fixed radius to cover all sensor nodes, and then to find a minimum cost tour passing all disk centers. The simulation results show the proposed algorithm outperforms the related CSP (Covering Salesman Problem) algorithm and the QiF algorithm.
{"title":"A Decreasing k-means algorithm for the Disk Covering Tour Problem in wireless sensor networks","authors":"Jia-Jiun Yang, Jehn-Ruey Jiang, Yung-Liang Lai","doi":"10.1109/PADSW.2014.7097906","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097906","url":null,"abstract":"This paper studies a Disk Covering Tour Problem (DCTP) for reducing the energy consumption of a mobile robot's movement to provide services for sensor nodes in a wireless sensor network (WSN). Given a set of locations of sensor nodes and a starting location of mobile robot, the DCTP is to find a minimum cost tour of a sequence of tour stops for the mobile robot to serve sensor nodes by keeping every sensor node within a specified distance of a tour stop. We propose an algorithm, called Decreasing k-means (Dk-means), to find an approximate solution to the DCTP. The idea is to select a minimum number of disks or circles of a fixed radius to cover all sensor nodes, and then to find a minimum cost tour passing all disk centers. The simulation results show the proposed algorithm outperforms the related CSP (Covering Salesman Problem) algorithm and the QiF algorithm.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134488898","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097829
Toshio Endo, Akira Nukada, S. Matsuoka
Modern supercomputer performance is principally limited by power. TSUBAME-KFC is a state-of-the-art prototype for our next-generation TSUBAME3.0 supercomputer and towards future exascale. In collaboration with Green Revolution Cooling and others, TSUBAME-KFC submerges compute nodes configured with extremely high processor/component density, into non-toxic, low viscosity oil with high 260 Celsius flash point, and cooled using ambient / evaporative cooling tower. This minimizes cooling power while all semiconductor components kept at low temperature to lower leakage current. Numerous off-line in addition to on-line power and temperature sensors are facilitated throughout and constantly monitored to immediately observe the effect of voltage/frequency control. As a result, TSUBAME-KFC achieved world No.1 on the Green500 in Nov. 2013 and Jun. 2014, by over 20% c.f. the nearest competitors.
{"title":"TSUBAME-KFC: A modern liquid submersion cooling prototype towards exascale becoming the greenest supercomputer in the world","authors":"Toshio Endo, Akira Nukada, S. Matsuoka","doi":"10.1109/PADSW.2014.7097829","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097829","url":null,"abstract":"Modern supercomputer performance is principally limited by power. TSUBAME-KFC is a state-of-the-art prototype for our next-generation TSUBAME3.0 supercomputer and towards future exascale. In collaboration with Green Revolution Cooling and others, TSUBAME-KFC submerges compute nodes configured with extremely high processor/component density, into non-toxic, low viscosity oil with high 260 Celsius flash point, and cooled using ambient / evaporative cooling tower. This minimizes cooling power while all semiconductor components kept at low temperature to lower leakage current. Numerous off-line in addition to on-line power and temperature sensors are facilitated throughout and constantly monitored to immediately observe the effect of voltage/frequency control. As a result, TSUBAME-KFC achieved world No.1 on the Green500 in Nov. 2013 and Jun. 2014, by over 20% c.f. the nearest competitors.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"60 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129581158","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097867
Umesh Chandra Satish, Praveenkumar Kondikoppa, Seung-Jong Park, Manish Patil, R. Shah
Genome indexing is the basis for many bioinformatics applications. Read mapping(sequence alignment) is one such application where the goal is to align millions of short reads against reference genome. Several tools are available for read mapping which rely on different indexing techniques to expedite the alignment process. However, many of these contemporary alignment programs are sequential, memory intensive and cannot be easily scaled for larger genomes. Suffix tree is one of the most widely used data structures for indexing strings (genomes). Building a scalable suffix-tree based tool is particularly challenging due to the difficulties involved in parallel construction of the suffix tree. Several suffix tree construction techniques have been proposed till date with focus on space-time tradeoff. Most of these existing works address the construction issue for uniprocessor and cannot be easily extended to utilize modern multi-processor systems. In this paper we investigate and propose a MapReduce based parallel construction of suffix tree. We demonstrate the performance of the algorithm over commodity cluster using up to 32 nodes each having 8GB of primary memory.
{"title":"MapReduce based parallel suffix tree construction for human genome","authors":"Umesh Chandra Satish, Praveenkumar Kondikoppa, Seung-Jong Park, Manish Patil, R. Shah","doi":"10.1109/PADSW.2014.7097867","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097867","url":null,"abstract":"Genome indexing is the basis for many bioinformatics applications. Read mapping(sequence alignment) is one such application where the goal is to align millions of short reads against reference genome. Several tools are available for read mapping which rely on different indexing techniques to expedite the alignment process. However, many of these contemporary alignment programs are sequential, memory intensive and cannot be easily scaled for larger genomes. Suffix tree is one of the most widely used data structures for indexing strings (genomes). Building a scalable suffix-tree based tool is particularly challenging due to the difficulties involved in parallel construction of the suffix tree. Several suffix tree construction techniques have been proposed till date with focus on space-time tradeoff. Most of these existing works address the construction issue for uniprocessor and cannot be easily extended to utilize modern multi-processor systems. In this paper we investigate and propose a MapReduce based parallel construction of suffix tree. We demonstrate the performance of the algorithm over commodity cluster using up to 32 nodes each having 8GB of primary memory.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"79 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132724583","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097902
Eduardo Cerritos, F. Lin
Uncertainty is a key factor that prevents a commuter from using public transportation system. More and more transportation agencies are incorporating real-time Trip Planners to empower commuters with opportune information. However, such systems require continuous status updates from the vehicles and involves expensive communication cost. In this paper we propose an architecture that takes advantage of Machine-to-Machine Communication concepts and provides a degree of intelligence to the vehicles, to alleviate unnecessary communication between the vehicles and the Trip Planner.
{"title":"M2M-enabled real-time Trip Planner","authors":"Eduardo Cerritos, F. Lin","doi":"10.1109/PADSW.2014.7097902","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097902","url":null,"abstract":"Uncertainty is a key factor that prevents a commuter from using public transportation system. More and more transportation agencies are incorporating real-time Trip Planners to empower commuters with opportune information. However, such systems require continuous status updates from the vehicles and involves expensive communication cost. In this paper we propose an architecture that takes advantage of Machine-to-Machine Communication concepts and provides a degree of intelligence to the vehicles, to alleviate unnecessary communication between the vehicles and the Trip Planner.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"228 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122461149","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097826
Xuedi Chen, Kenli Li, Chubo Liu, Kuan-Ching Li
The energy aware scheduling problem has been a critical issue in high-performance clusters owing to their high operation cost, environmental impact, and low reliability. An existing technique to reduce energy consumption of applications is dynamic voltage/frequency scaling (DVFS). In this paper, we develop an energy aware scheduling algorithm called EASLA for precedence-constrained applications in the context of Service Level Agreement (SLA) on DVFS-enabled cluster systems. Due to the dependencies among tasks and makespan extension, there may be some slacks under used. The main idea of the EASLA algorithm is to distribute each slack to a set of tasks and scale frequencies down to try to minimize energy consumption. Specifically, it first finds the maximum set of independent tasks for each task, and then iteratively allocates each slack to the maximum independent set whose total energy reduction is the maximal. Randomly generated graphs and two real-world applications are tested in our experiments. The experimental results show that our scheduling algorithm can save up to 22.68% and 12.01% energy consumption compared with GreedyDVS and EvenlyDVS algorithms, respectively.
{"title":"SLA-based energy aware scheduling of precedence-constrained applications on DVFS-enabled clusters","authors":"Xuedi Chen, Kenli Li, Chubo Liu, Kuan-Ching Li","doi":"10.1109/PADSW.2014.7097826","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097826","url":null,"abstract":"The energy aware scheduling problem has been a critical issue in high-performance clusters owing to their high operation cost, environmental impact, and low reliability. An existing technique to reduce energy consumption of applications is dynamic voltage/frequency scaling (DVFS). In this paper, we develop an energy aware scheduling algorithm called EASLA for precedence-constrained applications in the context of Service Level Agreement (SLA) on DVFS-enabled cluster systems. Due to the dependencies among tasks and makespan extension, there may be some slacks under used. The main idea of the EASLA algorithm is to distribute each slack to a set of tasks and scale frequencies down to try to minimize energy consumption. Specifically, it first finds the maximum set of independent tasks for each task, and then iteratively allocates each slack to the maximum independent set whose total energy reduction is the maximal. Randomly generated graphs and two real-world applications are tested in our experiments. The experimental results show that our scheduling algorithm can save up to 22.68% and 12.01% energy consumption compared with GreedyDVS and EvenlyDVS algorithms, respectively.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125126723","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 : 2014-12-01DOI: 10.1109/PADSW.2014.7097816
Ruiqing Chi, Zhuzhong Qian, Sanglu Lu
With the rapid development of virtualization techniques, modern data centers move into a new era of cloud in recent years. Despite numerous advantages such as high resource utilization and rapid service scalability, current virtualization techniques don't guarantee perfect performance isolation among virtual machines sharing the physical machine, which may lead to unstable and unpredictable user-perceived application performance in clouds. Therefore, understanding and modeling performance interference among collocated applications is of utmost importance. However, the hypervisor and guest OSes usually run independent resource schedulers and are invisible into each other, thereby making accurately characterizing performance interference a non-trivial work. In this paper, we first present a comprehensive experimental study on performance interference of different combinations of benchmarks, observing that virtual CPU floating overhead between multiple physical CPUs, and VMEXITs, i.e., the control transitions between the hypervisor and VMs, constitute the key source of performance interference. In order to characterize the performance interference effects, we measure both the application-level and VM-level characteristics from the collocated applications and then build a novel interference prediction framework based on kernel canonical correlation analysis. Our evaluations first show the practicability of KCCA in finding reliable correlation, and further confirm the high accuracy and great applicability of our interference model with a low prediction error of no more than 7.9%.
{"title":"Be a good neighbour: Characterizing performance interference of virtual machines under xen virtualization environments","authors":"Ruiqing Chi, Zhuzhong Qian, Sanglu Lu","doi":"10.1109/PADSW.2014.7097816","DOIUrl":"https://doi.org/10.1109/PADSW.2014.7097816","url":null,"abstract":"With the rapid development of virtualization techniques, modern data centers move into a new era of cloud in recent years. Despite numerous advantages such as high resource utilization and rapid service scalability, current virtualization techniques don't guarantee perfect performance isolation among virtual machines sharing the physical machine, which may lead to unstable and unpredictable user-perceived application performance in clouds. Therefore, understanding and modeling performance interference among collocated applications is of utmost importance. However, the hypervisor and guest OSes usually run independent resource schedulers and are invisible into each other, thereby making accurately characterizing performance interference a non-trivial work. In this paper, we first present a comprehensive experimental study on performance interference of different combinations of benchmarks, observing that virtual CPU floating overhead between multiple physical CPUs, and VMEXITs, i.e., the control transitions between the hypervisor and VMs, constitute the key source of performance interference. In order to characterize the performance interference effects, we measure both the application-level and VM-level characteristics from the collocated applications and then build a novel interference prediction framework based on kernel canonical correlation analysis. Our evaluations first show the practicability of KCCA in finding reliable correlation, and further confirm the high accuracy and great applicability of our interference model with a low prediction error of no more than 7.9%.","PeriodicalId":421740,"journal":{"name":"2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117175785","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：