R. Popescu, A. Staikopoulos, Peng Liu, Antonio Brogi, S. Clarke
Current adaptation approaches mainly work in isolation and cannot be easily integrated to tackle complex adaptation scenarios. The few existing cross-layer adaptation techniques are somewhat inflexible because the adaptation process is predefined and static. In this paper we propose a methodology for the dynamic and flexible adaptation of multi-layer applications. We use events to trigger the process of matching adaptation templates, which expose adaptation logic as BPEL processes. The matching process employs taxonomies of adaptation mismatches to select adaptation templates based on the degree of match between events and adaptation mismatches. We provide support for cross-layer adaptation by allowing templates to be composed either directly, through invocations of WSDL operations or indirectly, through events.
{"title":"Taxonomy-Driven Adaptation of Multi-layer Applications Using Templates","authors":"R. Popescu, A. Staikopoulos, Peng Liu, Antonio Brogi, S. Clarke","doi":"10.1109/SASO.2010.23","DOIUrl":"https://doi.org/10.1109/SASO.2010.23","url":null,"abstract":"Current adaptation approaches mainly work in isolation and cannot be easily integrated to tackle complex adaptation scenarios. The few existing cross-layer adaptation techniques are somewhat inflexible because the adaptation process is predefined and static. In this paper we propose a methodology for the dynamic and flexible adaptation of multi-layer applications. We use events to trigger the process of matching adaptation templates, which expose adaptation logic as BPEL processes. The matching process employs taxonomies of adaptation mismatches to select adaptation templates based on the degree of match between events and adaptation mismatches. We provide support for cross-layer adaptation by allowing templates to be composed either directly, through invocations of WSDL operations or indirectly, through events.","PeriodicalId":370044,"journal":{"name":"2010 Fourth IEEE International Conference on Self-Adaptive and Self-Organizing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129671955","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}
Many large-scale distributed systems have been built with great complexity to run Internet services. Due to the heterogeneity and dynamics of complex systems, it is very difficult to characterize their behavior precisely for system management. While we collect large amount of monitoring data from distributed systems as system observables, it is hard for us to interpret the data without constructing reasonable system models. Our previous work proposed algorithms to extract invariants from monitoring data to profile complex systems. However, such invariants are extracted between pair wise system measurements but not among multiple measurements. Based on minimal redundancy maximal relevance subset selection and least angle regression, this paper proposes an efficient algorithm to automatically extract overlay invariants from the layer of pair wise invariant networks. The overlay invariants link separated pair wise invariant subnets and enable us to support many system management tasks such as fault detection and capacity planning. Experimental results from synthetic data and real commercial systems are also included to demonstrate the effectiveness and efficiency of our algorithm.
{"title":"Extracting Overlay Invariants of Distributed Systems for Autonomic System Management","authors":"Hanhuai Shan, Guofei Jiang, K. Yoshihira","doi":"10.1109/SASO.2010.17","DOIUrl":"https://doi.org/10.1109/SASO.2010.17","url":null,"abstract":"Many large-scale distributed systems have been built with great complexity to run Internet services. Due to the heterogeneity and dynamics of complex systems, it is very difficult to characterize their behavior precisely for system management. While we collect large amount of monitoring data from distributed systems as system observables, it is hard for us to interpret the data without constructing reasonable system models. Our previous work proposed algorithms to extract invariants from monitoring data to profile complex systems. However, such invariants are extracted between pair wise system measurements but not among multiple measurements. Based on minimal redundancy maximal relevance subset selection and least angle regression, this paper proposes an efficient algorithm to automatically extract overlay invariants from the layer of pair wise invariant networks. The overlay invariants link separated pair wise invariant subnets and enable us to support many system management tasks such as fault detection and capacity planning. Experimental results from synthetic data and real commercial systems are also included to demonstrate the effectiveness and efficiency of our algorithm.","PeriodicalId":370044,"journal":{"name":"2010 Fourth IEEE International Conference on Self-Adaptive and Self-Organizing Systems","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122874841","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}
Total Order Broadcast (TOB) is a fundamental building block at the core of a number of strongly consistent, fault-tolerant replication schemes. While it is widely known that the performance of existing TOB algorithms varies greatly depending on the workload and deployment scenarios, the problem of how to forecast their performance in realistic settings is, at current date, still largely unexplored. In this paper we address this problem by exploring the possibility of leveraging on machine learning techniques for building, in a fully decentralized fashion, performance models of TOB protocols. Based on an extensive experimental study considering heterogeneous workloads and multiple TOB protocols, we assess the accuracy and efficiency of alternative machine learning methods including neural networks, support vector machines, and decision tree-based regression models. We propose two heuristics for the feature selection phase, that allow to reduce its execution time up to two orders of magnitude incurring in a very limited loss of prediction accuracy.
全顺序广播(Total Order Broadcast, TOB)是许多强一致性、容错复制方案的核心组成部分。虽然众所周知，现有TOB算法的性能会因工作负载和部署场景的不同而有很大差异，但如何在现实环境中预测其性能的问题，目前仍在很大程度上未被探索。在本文中，我们通过探索利用机器学习技术以完全分散的方式构建TOB协议性能模型的可能性来解决这个问题。基于一项广泛的实验研究，考虑了异构工作负载和多种TOB协议，我们评估了替代机器学习方法的准确性和效率，包括神经网络、支持向量机和基于决策树的回归模型。我们为特征选择阶段提出了两种启发式方法，可以在非常有限的预测精度损失的情况下将其执行时间减少两个数量级。
{"title":"A Machine Learning Approach to Performance Prediction of Total Order Broadcast Protocols","authors":"Maria Couceiro, P. Romano, L. Rodrigues","doi":"10.1109/SASO.2010.41","DOIUrl":"https://doi.org/10.1109/SASO.2010.41","url":null,"abstract":"Total Order Broadcast (TOB) is a fundamental building block at the core of a number of strongly consistent, fault-tolerant replication schemes. While it is widely known that the performance of existing TOB algorithms varies greatly depending on the workload and deployment scenarios, the problem of how to forecast their performance in realistic settings is, at current date, still largely unexplored. In this paper we address this problem by exploring the possibility of leveraging on machine learning techniques for building, in a fully decentralized fashion, performance models of TOB protocols. Based on an extensive experimental study considering heterogeneous workloads and multiple TOB protocols, we assess the accuracy and efficiency of alternative machine learning methods including neural networks, support vector machines, and decision tree-based regression models. We propose two heuristics for the feature selection phase, that allow to reduce its execution time up to two orders of magnitude incurring in a very limited loss of prediction accuracy.","PeriodicalId":370044,"journal":{"name":"2010 Fourth IEEE International Conference on Self-Adaptive and Self-Organizing Systems","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123622966","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}
One of the major problems in managing large-scale distributed systems is the prediction of the application performance. The complexity of the systems and the availability of monitored data have motivated the applicability of machine learning and other statistical techniques to induce performance models and forecast performance degradation problems. However, there is a stringent need for additional experimental and comparative studies, since there is no optimal method for all cases. In addition to a deeper comparison of different statistical techniques, studies lack on two important dimensions: resilience to transient failures of the statistical techniques, and diagnostic abilities. In this work, we address these issues, presenting three main contributions: first, we establish the capability of different statistical learning techniques for forecasting the resource needs of component-based distributed systems, second, we investigate an analysis engine that is more robust to false alarms, introducing a novel algorithm that augments the predictive power of statistical learning methods by combining them with a statistical test to identify trends in resources usage, third, we investigate the applicability of statistical tests for identifying the nature and cause of performance problems in component-based distributed systems.
{"title":"Statistical Approaches to Predicting and Diagnosing Performance Problems in Component-Based Distributed Systems: An Experimental Evaluation","authors":"S. Correa, Renato Cerqueira","doi":"10.1109/SASO.2010.32","DOIUrl":"https://doi.org/10.1109/SASO.2010.32","url":null,"abstract":"One of the major problems in managing large-scale distributed systems is the prediction of the application performance. The complexity of the systems and the availability of monitored data have motivated the applicability of machine learning and other statistical techniques to induce performance models and forecast performance degradation problems. However, there is a stringent need for additional experimental and comparative studies, since there is no optimal method for all cases. In addition to a deeper comparison of different statistical techniques, studies lack on two important dimensions: resilience to transient failures of the statistical techniques, and diagnostic abilities. In this work, we address these issues, presenting three main contributions: first, we establish the capability of different statistical learning techniques for forecasting the resource needs of component-based distributed systems, second, we investigate an analysis engine that is more robust to false alarms, introducing a novel algorithm that augments the predictive power of statistical learning methods by combining them with a statistical test to identify trends in resources usage, third, we investigate the applicability of statistical tests for identifying the nature and cause of performance problems in component-based distributed systems.","PeriodicalId":370044,"journal":{"name":"2010 Fourth IEEE International Conference on Self-Adaptive and Self-Organizing Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130130580","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}
Random scale-free overlay topologies provide a number of properties like for example high resilience against failures of random nodes, small (average) diameter as well as good expansion and congestion characteristics that make them interesting for the use in large-scale distributed systems. A number of these properties have been shown to be influenced by the exponent of their power law degree distribution. In this article, we present a distributed rewiring scheme that is suitable to effectuate random scale-free overlay topologies with an adjustable degree distribution exponent. The scheme uses a biased random walk strategy to sample new endpoints of edges being rewired and relies on an equilibrium model for scale-free networks. The bias of the random walk strategy can be tuned to produce random scale-free networks with arbitrary degree distribution exponents greater than two. We argue that the rewiring strategy can be implemented in a distributed fashion based on a node’s information about its immediate neighbors. We present both analytical arguments as well as results that have been obtained in simulations of the proposed protocol.
{"title":"Distributed Creation and Adaptation of Random Scale-Free Overlay Networks","authors":"Ingo Scholtes","doi":"10.1109/SASO.2010.45","DOIUrl":"https://doi.org/10.1109/SASO.2010.45","url":null,"abstract":"Random scale-free overlay topologies provide a number of properties like for example high resilience against failures of random nodes, small (average) diameter as well as good expansion and congestion characteristics that make them interesting for the use in large-scale distributed systems. A number of these properties have been shown to be influenced by the exponent of their power law degree distribution. In this article, we present a distributed rewiring scheme that is suitable to effectuate random scale-free overlay topologies with an adjustable degree distribution exponent. The scheme uses a biased random walk strategy to sample new endpoints of edges being rewired and relies on an equilibrium model for scale-free networks. The bias of the random walk strategy can be tuned to produce random scale-free networks with arbitrary degree distribution exponents greater than two. We argue that the rewiring strategy can be implemented in a distributed fashion based on a node’s information about its immediate neighbors. We present both analytical arguments as well as results that have been obtained in simulations of the proposed protocol.","PeriodicalId":370044,"journal":{"name":"2010 Fourth IEEE International Conference on Self-Adaptive and Self-Organizing Systems","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115298699","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：