Abstract:With energy consumption in high-performance computing clouds growing rapidly, energy saving has become an important topic. Virtualization provides opportunities to save energy by enabling one physical machine (PM) to host multiple virtual machines (VMs). Dynamic voltage and frequency scaling (DVFS) is another technology to reduce energy consumption. However, in heterogeneous cloud environments where DVFS may be applied at the chip level or the core level, it is a great challenge to combine these two technologies efficiently. On per-core DVFS servers, cloud managers should carefully determine VM placements to minimize performance interference. On full-chip DVFS servers, cloud managers further face the choice of whether to combine VMs with different characteristics to reduce performance interference or to combine VMs with similar characteristics to take better advantage of DVFS. This paper presents a novel mechanism combining a VM placement algorithm and a frequency scaling method. We formulate this VM placement problem as an integer programming (IP) to find appropriate placement configurations, and we utilize support vector machines to select suitable frequencies. We conduct detailed experiments and simulations, showing that our scheme effectively reduces energy consumption with modest impact on performance. Particularly, the total energy delay product is reduced by up to 60%.

A group management to balance power,performance,and fairness

Time-sharing Parallel Applications Through Performance-Targeted Feedback-Controlled Real-Time Scheduling.

A Self-aware Dynamic Power Management System

Power Aware Job Scheduling with Quality of Service Guarantees : A Preliminary Study

Group Scheduling for Improving Both CPU and Memory Power Efficiency Simultaneously

Power-Aware Buddy System and Task Group Scheduler

A Dynamic Frequency Governor for Operating System Based on Performance-Energy Tradeoff Model

Coordinate Task and Memory Management for Improving Power Efficiency

Coordinate page allocation and thread group for improving main memory power efficiency

Dynamic power management of complex system using flow chart and poisson process

Power Management in Heterogeneous Multi-tier Web Clusters

A Power and Performance Management Framework for Virtualized Server Clusters

Analysis of Power Saving Effect for Dynamic Power Management

Balancing Power and Performance in HPC Clouds.

Memory Power Optimization Policy of Coordinating Page Allocation and Group Scheduling

Adaptive Performance Optimization under Power Constraint in Multi-thread Applications with Diverse Scalability

Dynamic Power Management and Dynamic Voltage Scaling in Real-Time Cmp Systems

A Criticality-Aware DVFS Runtime Utility for Optimizing Power Efficiency of Multithreaded Applications

The modeling of power management for complex system using normal distribution

Study of a low power optimization method based on multithreaded load imbalance for embedded multicore

Memory Affinity: Balancing Performance, Power, Thermal and Fairness for Multi-core Systems