The Challenge of Cloud control. Maria Kihl, Erik Elmroth, Johan Tordsson, Anders Robertsson, Karl Erik Årzén - PDF

Description
The Challenge of Cloud control Maria Kihl, Erik Elmroth, Johan Tordsson, Anders Robertsson, Karl Erik Årzén Vision of cloud data centers Self-managed, dynamic, and dependable infrastructure, constantly

Please download to get full document.

View again

of 17
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Games & Puzzles

Publish on:

Views: 9 | Pages: 17

Extension: PDF | Download: 0

Share
Transcript
The Challenge of Cloud control Maria Kihl, Erik Elmroth, Johan Tordsson, Anders Robertsson, Karl Erik Årzén Vision of cloud data centers Self-managed, dynamic, and dependable infrastructure, constantly delivering expected QoS with reasonable operation costs and acceptable carbon footprint for large-scale services with sometimes dramatic variations in capacity demands. Very basic concept of clouds Services Virtual machines Requests Infrastructure Challenges of Cloud Control Service Admission Control Services Holistic management control Elasticity control Placement control Energy optimization and control Virtual machines Physical machines Power & Cooling Project Cloud Control Large Framework grant from the Swedish Research Council. Cloud Research group at Umeå University (Prof. Erik Elmroth, Assoc. Prof Johan Tordsson, Prof. Bo Kågström) Lund Center for Control of Complex Engineering Systems (Prof. Karl Erik Årzén, Prof. Anders Rantzer, Prof. Anders Robertsson, Assoc. Prof. Maria Kihl) Lund Center for Control of Complex Engineering Systems (LCCC) Research center at Lund University, Sweden, funded by the Swedich Research Council. LCCC is developing theory, methods and tools for control of large-scale engineering systems. Main research areas: Distributed Decision-Making and Control Implementation in Networked and Embedded Systems Modeling Support for Design and Verification Read more: LCCC Research on Networked and Embedded Systems In LCCC, we have a high competence in control of computing systems. Performance modelling and load control. Prediction based and stochastic control mechanisms. Adaptive resource management for embedded systems. Experimental validation and testbeds. Umeå University The Umeå University distributed systems' research focuses on flexible and scalable IT infrastructures, and covers a range of topics central to cloud and grid computing. + several senior researchers, postdocs and PhD students Web: cloudresearch.se UU competence The group focuses on systems-oriented research on cloud management with industrial partners like IBM, SAP, ATOS, Telefonica, and British Telecom. Their competence is based on profound understanding of the challenges of future cloud infrastructures and the technology required to make them reality. The Umeå group is central to three of Europe s leading industry-driven cloud research projects. The team includes the director for the High Performance Computing Center North (HPC2N). Main vision of project Cross-disciplinary project based on the perfect match of complementing expertise on cloud management and control of computing systems. A control theoretic approach to a range of cloud management problems. Main aim to transform today s static and energy consuming cloud data centers into self-managed, dynamic, and dependable infrastructures, constantly delivering expected quality of service with acceptable operation costs and carbon footprint for large-scale services with varying capacity demands. Example: Performance models Performance models Example: Computing systems with CPUintensive workload can be modelled as single server queues (M/M/1 dynamics). l m Challenges for clouds Traffic performance monitoring of large distributed systems Workload models Scalability effects Example: Stochastic elasticity control Accurate models needed! Other topics placement control Optimization of the mapping of virtual resources to the infrastructure Data center energy optimization Holistic management How do different optimization and control mechanisms interact with each other? A global control system that monitors the complete system behavior and optimizes the local management systems. Testbeds and experiments One important goal is to validate theories in experiments. Available testbeds: Umeå s cloud infrastructure (40+ servers) Green server farm at University of Illinois at Urbana-Champaign Energy monitoring! Small cloud testbed at Lund University Focus period on Cloud Control LCCC will host a focus period and workshop on Cloud Control in April/May Some planned topics: Management and control of cloud infrastructures (e.g. elasticity control) Heat and power control in data centers Distributed Mobile Clouds More information will follow...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks