Proceedings of the10th International Conference on Utility and Cloud Computing最新文献_第2页

Session details: Energy and Green Data Center 会议详情:能源和绿色数据中心

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3258567

Richard Hill

引用次数: 0

Adaptive Service Performance Control using Cooperative Fuzzy Reinforcement Learning in Virtualized Environments 虚拟环境下基于协同模糊强化学习的自适应服务性能控制

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3147213.3147225

Olumuyiwa Ibidunmoye, M. H. Moghadam, Ewnetu Bayuh Lakew, E. Elmroth

{"title":"Adaptive Service Performance Control using Cooperative Fuzzy Reinforcement Learning in Virtualized Environments","authors":"Olumuyiwa Ibidunmoye, M. H. Moghadam, Ewnetu Bayuh Lakew, E. Elmroth","doi":"10.1145/3147213.3147225","DOIUrl":"https://doi.org/10.1145/3147213.3147225","url":null,"abstract":"Designing efficient control mechanisms to meet strict performance requirements with respect to changing workload demands without sacrificing resource efficiency remains a challenge in cloud infrastructures. A popular approach is fine-grained resource provisioning via auto-scaling mechanisms that rely on either threshold-based adaptation rules or sophisticated queuing/control-theoretic models. While it is difficult at design time to specify optimal threshold rules, it is even more challenging inferring precise performance models for the multitude of services. Recently, reinforcement learning have been applied to address this challenge. However, such approaches require many learning trials to stabilize at the beginning and when operational conditions vary thereby limiting their application under dynamic workloads. To this end, we extend the standard reinforcement learning approach in two ways: a) we formulate the system state as a fuzzy space and b) exploit a set of cooperative agents to explore multiple fuzzy states in parallel to speed up learning. Through multiple experiments on a real virtualized testbed, we demonstrate that our approach converges quickly, meets performance targets at high efficiency without explicit service models.","PeriodicalId":341011,"journal":{"name":"Proceedings of the10th International Conference on Utility and Cloud Computing","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132194293","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: 8

Session details: Networking and Storage 会话详细信息:网络和存储

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3258568

Xiaoyi Lu

引用次数: 0

Session details: Containerized Environments and Applications 会话详细信息:容器化环境和应用程序

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3258570

A. Anjum

引用次数: 0

Exploiting Efficiency Opportunities Based on Workloads with Electron on Heterogeneous Clusters 利用异构集群上基于电子负载的效率机会

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3147213.3147226

Renan Delvalle, Pradyumna Kaushik, Abhishek Jain, Jessica Hartog, M. Govindaraju

{"title":"Exploiting Efficiency Opportunities Based on Workloads with Electron on Heterogeneous Clusters","authors":"Renan Delvalle, Pradyumna Kaushik, Abhishek Jain, Jessica Hartog, M. Govindaraju","doi":"10.1145/3147213.3147226","DOIUrl":"https://doi.org/10.1145/3147213.3147226","url":null,"abstract":"Resource Management tools for large-scale clusters and data centers typically schedule resources based on task requirements specified in terms of processor, memory, and disk space. As these systems scale, two non-traditional resources also emerge as limiting factors: power and energy. Maintaining a low power envelope is especially important during Coincidence Peak, a window of time where power may cost up to 200 times the base rate. Using Electron, our power-aware framework that leverages Apache Mesos as a resource broker, we quantify the impact of four scheduling policies on three workloads of varying power intensity. We also quantify the impact of two dynamic power capping strategies on power consumption, energy consumption, and makespan when used in combination with scheduling policies across workloads. Our experiments show that choosing the right combination of scheduling and power capping policies can lead to a 16% reduction of energy and a 37% reduction in the 99th percentile of power consumption while having a negligible impact on makespan and resource utilization.","PeriodicalId":341011,"journal":{"name":"Proceedings of the10th International Conference on Utility and Cloud Computing","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124916528","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

Session details: Multi-Cloud Environments and Data Dissemination 会议详情:多云环境和数据传播

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3258566

Yanfei Guo

引用次数: 0

Session details: Edge Computing and Edge Cloud 会话详细信息:边缘计算和边缘云

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3258569

T. Cockerill

引用次数: 0

Dynamic Network Scheduler for Cloud Data Centres with SDN 基于SDN的云数据中心动态网络调度程序

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3147213.3147217

Christopher B. Hauser, Santhosh Ramapuram Palanivel

{"title":"Dynamic Network Scheduler for Cloud Data Centres with SDN","authors":"Christopher B. Hauser, Santhosh Ramapuram Palanivel","doi":"10.1145/3147213.3147217","DOIUrl":"https://doi.org/10.1145/3147213.3147217","url":null,"abstract":"The presented dynamic network scheduler improves the fairness and efficiency of network utilization in a Cloud data centre. The proposed design utilizes a directed graph which represents the network comprising routers, switches, physical hypervisors, and virtual machines (VMs) as graph nodes, and represents the physical network connections as weighted edges. The edges have a guaranteed transmission rate derived from the number of devices sharing an outgoing link with a defined bandwidth. Moreover, to maximize utilization of the resources, each node gets a deserved rate dynamically depending on the measured utilization metrics. A VM throttles up or down its traffic up to its deserved rate. The conceptual design of a dynamic network scheduler is further prototypically implemented and evaluated. The implementation uses Software Defined Networking (SDN) with OpenFlow, Ryu SDN controller, and Open vSwitch as software switch on the hypervisor level. The presented dynamic network scheduler uses OpenFlow for monitoring and applying flows to control the link bandwidth. The evaluation shows that the dynamic network scheduler maximizes fairness in resource sharing while minimizing the unutilized resources.","PeriodicalId":341011,"journal":{"name":"Proceedings of the10th International Conference on Utility and Cloud Computing","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122345460","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: 8

NIST Big Data Reference Architecture for Analytics and Beyond NIST大数据分析及其他参考架构

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3147213.3155013

Wo L. Chang

{"title":"NIST Big Data Reference Architecture for Analytics and Beyond","authors":"Wo L. Chang","doi":"10.1145/3147213.3155013","DOIUrl":"https://doi.org/10.1145/3147213.3155013","url":null,"abstract":"Big Data is the term used to describe the deluge of data in our networked, digitized, sensor-laden, information driven world. There is a broad agreement among commercial, academic, and government leaders about the remarkable potential of \"Big Data\" to spark innovation, fuel commerce, and drive progress. The availability of vast data resources carries the potential to answer questions previously out of reach. However, there is also broad agreement on the ability of Big Data to overwhelm traditional approaches. Big Data architectures come in many shapes and forms ranging from academic research settings to product-oriented workflows. With massive-scale dynamic data being generate from social media, Internet of Things, Smart Cities, and others, it is critical to analyze these data in real-time and provide proactive decision. With the advancement of computer architecture in multi-cores and GPUs, and fast communication between CPUs and GPUs, parallel processing utilizes these platforms could optimize resources at a reduced time. This presentation will provide the past, current, and future activities of the NIST Big Data Public Working Group (NBD-PWG) and how the NIST Reference Architecture may address the rate at which data volumes, speeds, and complexity are growing requires new forms of computing infrastructure to enable Big Data analytics interoperability such that analytics tools can be re-usable, deployable, and operational. The focus of NBD-PWG is to form a community of interest from industry, academia, and government, with the goal of developing consensus definitions, taxonomies, secure reference architectures, and standards roadmap which would create vendor-neutral, technology and infrastructure agnostic framework. The aim is to enable Big Data stakeholders to pick-and-choose best analytics tools for their processing under the most suitable computing platforms and clusters while allowing value-additions from Big Data service providers and flow of data between the stakeholders in a cohesive and secure manner.","PeriodicalId":341011,"journal":{"name":"Proceedings of the10th International Conference on Utility and Cloud Computing","volume":"204 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-12-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131463582","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: 3

Session details: Scheduling and Resource Management 会话详细信息:调度和资源管理

Proceedings of the10th International Conference on Utility and Cloud Computing Pub Date : 2017-12-05 DOI: 10.1145/3258565

Dong Dai

引用次数: 0