2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)最新文献_第2页

Hierarchical and Frequency-Aware Model Predictive Control for Bare-Metal Cloud Applications 裸机云应用的层次和频率感知模型预测控制

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00010

Yukio Ogawa, G. Hasegawa, M. Murata

{"title":"Hierarchical and Frequency-Aware Model Predictive Control for Bare-Metal Cloud Applications","authors":"Yukio Ogawa, G. Hasegawa, M. Murata","doi":"10.1109/UCC.2018.00010","DOIUrl":"https://doi.org/10.1109/UCC.2018.00010","url":null,"abstract":"Bare-metal cloud provides a dedicated set of physical machines (PMs) and enables both PMs and virtual machines (VMs) on the PMs to be scaled in/out dynamically. However, to increase efficiency of the resources and reduce violations of service level agreements (SLAs), resources need to be scaled quickly to adapt to workload changes, which results in high reconfiguration overhead, especially for the PMs. This paper proposes a hierarchical and frequency-aware auto-scaling based on Model Predictive Control, which enable us to achieve an optimal balance between resource efficiency and overhead. Moreover, when performing high-frequency resource control, the proposed technique improves the timing of reconfigurations for the PMs without increasing the number of them, while it increases the reallocations for the VMs to adjust the redundant capacity among the applications; this process improves the resource efficiency. Through trace-based numerical simulations, we demonstrate that when the control frequency is increased to 16 times per hour, the VM insufficiency causing SLA violations is reduced to a minimum of 0.1% per application without increasing the VM pool capacity.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131876013","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}

引用次数: 1

Return on Investment for Three Cyberinfrastructure Facilities: A Local Campus Supercomputer, the NSF-Funded Jetstream Cloud System, and XSEDE (the eXtreme Science and Engineering Discovery Environment) 三个网络基础设施的投资回报:本地校园超级计算机、nsf资助的Jetstream云系统和XSEDE(极端科学与工程发现环境)

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00031

C. Stewart, David Y. Hancock, Julie A. Wernert, Matthew R. Link, Nancy Wilkins-Diehr, Therese Miller, K. Gaither, Winona Snapp-Childs

{"title":"Return on Investment for Three Cyberinfrastructure Facilities: A Local Campus Supercomputer, the NSF-Funded Jetstream Cloud System, and XSEDE (the eXtreme Science and Engineering Discovery Environment)","authors":"C. Stewart, David Y. Hancock, Julie A. Wernert, Matthew R. Link, Nancy Wilkins-Diehr, Therese Miller, K. Gaither, Winona Snapp-Childs","doi":"10.1109/UCC.2018.00031","DOIUrl":"https://doi.org/10.1109/UCC.2018.00031","url":null,"abstract":"The economics of high performance computing are rapidly changing. Commercial cloud offerings, private research clouds, and pressure on the budgets of institutions of higher education and federally-funded research organizations are all contributing factors. As such, it has become a necessity that all expenses and investments be analyzed and considered carefully. In this paper we will analyze the return on investment (ROI) for three different kinds of cyberinfrastructure resources: the eXtreme Science and Engineering Discovery Environment (XSEDE); the NSF-funded Jetstream cloud system; and the Indiana University (IU) Big Red II supercomputer, funded exclusively by IU for use of the IU community and collaborators. We determined the ROI for these three resources by assigning financial values to services by either comparison with commercially available services, or by surveys of value of these resources to their users. In all three cases, the ROI for these very different types of cyberinfrastructure resources was well greater than 1 - meaning that investors are getting more than $1 in returned value for every $1 invested. While there are many ways to measure the value and impact of investment in cyberinfrastructure resources, we are able to quantify the short-term ROI and show that it is a net positive for campuses and the federal government respectively.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128294219","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}

引用次数: 11

QoS Aware Energy Efficient VM Consolidation Techniques for a Virtualized Data Center 基于QoS的虚拟化数据中心虚拟机节能整合技术

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00020

Anurina Tarafdar, Sunirmal Khatua, R. Das

{"title":"QoS Aware Energy Efficient VM Consolidation Techniques for a Virtualized Data Center","authors":"Anurina Tarafdar, Sunirmal Khatua, R. Das","doi":"10.1109/UCC.2018.00020","DOIUrl":"https://doi.org/10.1109/UCC.2018.00020","url":null,"abstract":"The large-scale virtualized data centers in the Cloud environment consume huge amount of energy leading to high operational costs and emission of greenhouse gases. Energy consumption of a data center can be reduced by dynamically consolidating the virtual machines (VMs) to a minimum number of physical machines, using live migration. However, the dynamic workload of virtual machines makes the VM consolidation problem more challenging. In this paper, we have proposed a prediction based migration technique for the VMs, where we perform VM migrations based on the predicted CPU utilization. Extensive simulations show that the proposed technique substantially reduces energy consumption, number of VM migrations and Service Level Agreement (SLA) violations within a data center. The performance overheads associated with excessive migration of VMs increase the time needed by the VMs to complete their jobs. So in this paper, we have also proposed a deadline aware VM migration technique, which reduces the time taken by the VMs to execute their jobs significantly, thereby improving the Quality of Service (QoS). Such improvement in QoS is achieved at the cost of slight increase in the energy consumption within the data center. However, simulation results show that appropriate setting of deadlines for the VMs, helps in achieving a trade-off between energy consumption and the QoS.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133288099","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}

引用次数: 4

Pacer: Automated Feedback-Based Vertical Elasticity for Heterogeneous Soft Real-Time Workloads 基于自动反馈的异构软实时工作负载垂直弹性

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00016

Yu-An Chen, Geoffrey Phi C. Tran, A. Rittenbach, J. Walters, S. Crago

引用次数: 0

UCC 2018 Organizing Committee UCC 2018组委会

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/ucc.2018.00007

引用次数: 1

Profiling and Predicting Application Performance on the Cloud 分析和预测云上的应用程序性能

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00011

Matt Baughman, Ryan Chard, Logan T. Ward, Jason Pitt, K. Chard, Ian T Foster

{"title":"Profiling and Predicting Application Performance on the Cloud","authors":"Matt Baughman, Ryan Chard, Logan T. Ward, Jason Pitt, K. Chard, Ian T Foster","doi":"10.1109/UCC.2018.00011","DOIUrl":"https://doi.org/10.1109/UCC.2018.00011","url":null,"abstract":"Cloud providers continue to expand and diversify their collection of leasable resources to meet the needs of an increasingly wide range of applications. While this flexibility is a key benefit of the cloud, it also creates a complex landscape in which users are faced with many resource choices for a given application. Suboptimal selections can both degrade performance and increase costs. Given the rapidly evolving pool of resources, it is infeasible for users alone to select instance types; instead, automated methods are needed to simplify and guide resource provisioning. Here we present a method for the automatic prediction of application performance on arbitrary cloud instances. We combine offline and online profiling approaches, using historical data gathered from non-cloud environments and targeted profiling runs on cloud environments to create a composite application model that can predict run times on a given cloud instance type for a given input data size. We demonstrate average error of 17.2% across nine applications used in production bioinformatics workflows. Finally, we evaluate an experiment design approach to explore the trade-off between the cost of profiling and the accuracy of our models. Using this approach, with no prior knowledge, we show that using 4 selectively chosen experiments we can achieve performance within 30% of a model trained using all instance types.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"47 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124019422","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}

引用次数: 15

Tromino: Demand and DRF Aware Multi-Tenant Queue Manager for Apache Mesos Cluster Tromino:用于Apache Mesos集群的需求和DRF感知的多租户队列管理器

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00015

Pankaj Saha, Angel Beltre, M. Govindaraju

{"title":"Tromino: Demand and DRF Aware Multi-Tenant Queue Manager for Apache Mesos Cluster","authors":"Pankaj Saha, Angel Beltre, M. Govindaraju","doi":"10.1109/UCC.2018.00015","DOIUrl":"https://doi.org/10.1109/UCC.2018.00015","url":null,"abstract":"Apache Mesos, a two-level resource scheduler, provides resource sharing across multiple users in a multi-tenant clustered environment. Computational resources (i.e., CPU, memory, disk, etc.) are distributed according to the Dominant Resource Fairness (DRF) policy. Mesos frameworks (users) receive resources based on their current usage and are responsible for scheduling their tasks within the allocation. We have observed that multiple frameworks can cause fairness imbalance in a multi-user environment. For example, a greedy framework consuming more than its fair share of resources can deny resource fairness to others. The user with the least Dominant Share is considered first by the DRF module to get its resource allocation. However, the default DRF implementation, in Apache Mesos' Master allocation module, does not consider the overall resource demands of the tasks in the queue for each user/framework. This lack of awareness can lead to poor performance as users without any pending task may receive more resource offers, and users with a queue of pending tasks can starve due to their high dominant shares. In a multi-tenant environment, the characteristics of frameworks and workloads must be understood by cluster managers to be able to define fairness based on not only resource share but also resource demand and queue wait time. We have developed a policy driven queue manager, Tromino, for an Apache Mesos cluster where tasks for individual frameworks can be scheduled based on each framework's overall resource demands and current resource consumption. Dominant Share and demand awareness of Tromino and scheduling based on these attributes can reduce (1) the impact of unfairness due to a framework specific configuration, and (2) unfair waiting time due to higher resource demand in a pending task queue. In the best case, Tromino can significantly reduce the average waiting time of a framework by using the proposed Demand-DRF aware policy.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127497887","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}

引用次数: 5

Pattern-Based Deployment Models and Their Automatic Execution 基于模式的部署模型及其自动执行

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00013

Lukas Harzenetter, Uwe Breitenbücher, Michael Falkenthal, Jasmin Guth, Christoph Krieger, F. Leymann

{"title":"Pattern-Based Deployment Models and Their Automatic Execution","authors":"Lukas Harzenetter, Uwe Breitenbücher, Michael Falkenthal, Jasmin Guth, Christoph Krieger, F. Leymann","doi":"10.1109/UCC.2018.00013","DOIUrl":"https://doi.org/10.1109/UCC.2018.00013","url":null,"abstract":"The automated deployment of cloud applications is of vital importance. Therefore, several deployment automation technologies have been developed that enable automatically deploying applications by processing so-called deployment models, which describe the components and relationships an application consists of. However, the creation of such deployment models requires considerable expertise about the technologies and cloud providers used—especially for the technical realization of conceptual architectural decisions. Moreover, deployment models have to be adapted manually if architectural decisions change or technologies need to be replaced, which is time-consuming, error-prone, and requires even more expertise. In this paper, we tackle this issue. We introduce a meta-model for Pattern-based Deployment Models, which enables using cloud patterns as generic, vendor-, and technology-agnostic modeling elements directly in deployment models. Thus, instead of specifying concrete technologies, providers, and their configurations, our approach enables modeling only the abstract concepts represented by patterns that must be adhered to during the deployment. Moreover, we present how these models can be automatically refined to executable deployment models. To validate the practical feasibility of our approach, we present a prototype based on the TOSCA standard and a case study.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124726342","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}

引用次数: 12

Two Efficient QoS-Based Approaches for a Resource Splitting Strategy across Multiple Cloud Providers 跨多个云提供商的资源分割策略的两种基于qos的有效方法

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00026

Marieme Diallo, A. Quintero, S. Pierre

{"title":"Two Efficient QoS-Based Approaches for a Resource Splitting Strategy across Multiple Cloud Providers","authors":"Marieme Diallo, A. Quintero, S. Pierre","doi":"10.1109/UCC.2018.00026","DOIUrl":"https://doi.org/10.1109/UCC.2018.00026","url":null,"abstract":"In this paper, we address the problem of com-putational and networking virtual resources embedding across multiple Infrastructure-as-a-Service (IaaS) providers. This issue, usually referred to as the Virtual Network Embedding (VNE) problem, requires two phases of operation in such a context: the multicloud virtual network requests (VNRs) splitting, followed by the intracloud VNR segments mapping. This paper focuses on the splitting phase problem, by proposing a splitting strategy based on two optimization approaches, with the objective of improving the performance and the quality of service (QoS) of resulting mapped VNR segments. An Integer Linear Program (ILP) is used to formalize our splitting strategy as a mathematical minimization problem with constraints. The ILP model is first solved with the exact approach. Subsequently, a metaheuristic approach based on the Tabu Search (TS) is proposed in order to find optimal or near-optimal solutions in polynomial solving time. The simulation results obtained show the efficiency of the proposed VNRs splitting approaches according to several performance criteria. Solution costs of the heuristic are on average close to the exact solution, with an average cost gap ranging from 0% to a maximum of 2.05%, performed in a highly reduced computing time. In comparison with other baseline approaches, the acceptance rate and the delay are improved by approximately 15%, while preventing QoS violations.","PeriodicalId":288232,"journal":{"name":"2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129636028","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}

引用次数: 0

A Multi-Cloud Marketplace Model with Multiple Brokers for IaaS Layer and Generalized Stable Matching 面向IaaS层的多代理多云市场模型及广义稳定匹配

2018 IEEE/ACM 11th International Conference on Utility and Cloud Computing (UCC) Pub Date : 2018-12-01 DOI: 10.1109/UCC.2018.00034

Shrenik Jain, Suresh Purini, P. V. Reddy

引用次数: 6