Abstract
This paper explores the feasibility of and challenges in developing methods for black-box monitoring of the power usage of a virtual machine (VM) at run-time, on shared virtualized compute platforms, including those with complex memory hierarchies. We demonstrate that VM-level power utilization can be accurately estimated, or estimated with accuracy with bound error margins. The use of bounds permits more lightweight online monitoring of fewer events, while relaxing the fidelity of the estimates in a controlled manner. Our methodology is evaluated on the Intel Core i7 and Core2 x86-64 platforms, running synthetic and SPEC benchmarks.
- BOHRA, A., AND CHAUDHARY, V. Vmeter: Power Modelling for Virtualized Clouds. In IPDPS (2010).Google Scholar
- CARTER, J. Looming Challenges in Server and Data Center Energy Efficiency, One Researcher's Perspective. CERCS Energy Workshop. 2010.Google Scholar
- CHOU, Y., FAHS, B., AND ABRAHAM, S. Microarchitecture Optimizations for Exploiting Memory-Level Parallelism. In HPCA (2005).Google Scholar
- HP Data Center Management Solution Reduces Costs by 34 Percent. www.hp.com/hpinfo/newsroom/press/2007/070625xa.html.Google Scholar
- IBM Helps Clients "Meter" Datacenter Power Usage to Help Lower Energy Costs. www-03.ibm.com/press/us/en/pressrelease/19695.wss.Google Scholar
- IBM, Syracuse University, New York State to Build One of the World's Most Energy-Efficient Data Centers. www-03.ibm.com/press/us/en/pressrelease/27612.wss.Google Scholar
- KANSAL, A., KOTHARI, N., AND BHATTACHARYA, A. Virtual Machine Power Metering and Provisioning. In ACM SOCC (2010). Google ScholarDigital Library
- KOLLER, R., VERMA, A., AND NEOGI, A. WattApp: An Application Aware Power Meter for Shared Data Centers. In ICAC (2010). Google ScholarDigital Library
- KUMAR, S., TALWAR, V., ET AL. vManage: Loosely-coupled Platform and Virtualization Management in Datacenters. In ICAC (2009). Google ScholarDigital Library
- KUSIC, D., KEPHART, J., ET AL. Power and Performance Management of Virtualized Computing Environments via Lookahead Control. In ICAC (2008). Google ScholarDigital Library
- NATHUJI, R., AND KANSAL, A. Q-Clouds: Managing Performance. Interference Effects for QoS-Aware Clouds. In Eurosys (2010). Google ScholarDigital Library
- NATHUJI, R., AND SCHWAN, K. VPMTokens: Virtual Machine-Aware Power Budgeting in Datacenters. In HPDC (2008). Google ScholarDigital Library
- NATHUJI, R., SOMANI, A., SCHWAN, K., AND JOSHI, Y. CoolIT: Coordinating Facility and IT Management for Efficient Datacenters. In USENIX HotPower (2008). Google ScholarDigital Library
- SAMADIANI, E., AMUR, H., KRISHNAN, B., AND SCHWAN, K. Coordinated Optimization of Cooling and IT Power in Data Centers. ASME Journal of Electronics Packaging (2010).Google Scholar
- STOESS, J., LANG, C., AND BELLOSA, F. Energy Management for Hypervisor-based Virtual Machines. In USENIX (2007). Google ScholarDigital Library
- STRONG, B. A Look Inside Intel: The Core (Nehalem) Microarchitecture.Google Scholar
Index Terms
- VM power metering: feasibility and challenges
Recommendations
Process-level power estimation in VM-based systems
EuroSys '15: Proceedings of the Tenth European Conference on Computer SystemsPower estimation of software processes provides critical indicators to drive scheduling or power capping heuristics. State-of-the-art solutions can perform coarse-grained power estimation in virtualized environments, typically treating virtual machines (...
Virtual machine power metering and provisioning
SoCC '10: Proceedings of the 1st ACM symposium on Cloud computingVirtualization is often used in cloud computing platforms for its several advantages in efficiently managing resources. However, virtualization raises certain additional challenges, and one of them is lack of power metering for virtual machines (VMs). ...
Network-centric Performance Improvement for Live VM Migration
CLOUD '15: Proceedings of the 2015 IEEE 8th International Conference on Cloud ComputingLive Virtual Machine (VM) migrations are an important tool that is used in modern data centers in order to e.g. Consolidate server racks for maintenance or optimize VM placements across physical hosts. However, live VM migration causes a lot of network ...
Comments