Developing resource consolidation frameworks for moldable virtual machines

Developing resource consolidation frameworks for moldable virtual machines in clouds Author: Liang He, Deqing Zou, Zhang, etc Presenter: Weida Zhong

abstract • Using Genetic Algorithm to consolidate moldable Virtual Machines • Developing a reconfiguration algorithm to lower the transition overhead that transiting the Cloud to the optimized system state needs

contents • System Hierarchy and workload models • Genetic Algorithm • Reconfiguring virtual clusters 1. Categorizing changes in system state 2. Transiting system states 3. Calculating transition time • Experimental studies

System Hierarchy and workload models

System Hierarchy and workload models • The cloud system aims to maintain a steady level of Quality of Service (Qo. S) delivered by every VC. • The desired Qo. S is expressed as that the total service rate of all VMs in a VC cannot be less than a certain figure

contents • System Hierarchy and workload models • Genetic Algorithm • Reconfiguring virtual clusters 1. Categorizing changes in system state 2. Transiting system states 3. Calculating transition time • Experimental studies

Classical genetic algorithm procedure begin initialization Next generation Evaluation/ fitness computing mutation crossover No Stop? end Yes reproduction

Genetic Algorithm •

Schematic diagram Current active new

Genetic Algorithm •

Genetic Algorithm -- crossover •

Genetic Algorithm -- mutation • 1. determining index i, j, k Select VC Select Node Select Resource The ratio of the probability of selecting the major resource type to other resource types is set to be R : 1 (R is the number of resource types in the system)

Genetic Algorithm -- mutation •

Genetic Algorithm – fitness function •

contents • System Hierarchy and workload models • Genetic Algorithm • Reconfiguring virtual clusters 1. Categorizing changes in system state 2. Transiting system states 3. Calculating transition time • Experimental studies

Reconfiguring virtual clusters S GA cost? Changing Capacity VM Creation VM Deletion VM Migration

Categorizing changes in system state •

Transiting system states ---VM operations during the transition • number of request average execution time of a request duration that the current request has been run

Transiting system states --- VM operations during the transition • releasing allocating

Transiting system states ---Performing VM operations without dependency •

Transiting system states ---Performing VM operations with dependency •

Algorithm 2(cont. ) •

Algorithm 4. Reconfiguring the cloud •

Calculating transition time • A Directed Acyclic Graph (DAG) can be constructed based on the dependencies between the VM operations as well as between source nodes and mapping destination nodes.

Calculating transition time • If the VM operations in all nodes form a single DAG, calculating the transition time of the reconfiguration plan for the cloud can be transformed to compute the critical path in the DAG. • If there are several DAG graphs, the time of the longest critical path is the transition time of the reconfiguration plan

contents • System Hierarchy and workload models • Genetic Algorithm • Reconfiguring virtual clusters 1. Categorizing changes in system state 2. Transiting system states 3. Calculating transition time • Experimental studies

Experimental studies • The simulation experiments about the effectiveness of the GA algorithm • The performance of the cloud reconfiguration method

Performance of GA --- impact of the number of physical nodes

Performance of GA --- impact of free capacity

Performance of GA --- impact of the number of VCs

Performance of the cloud reconfiguration

Conclusion • Develop a resource consolidation framework for moldable virtual machines in clouds • A Genetic Algorithm is developed to compute the optimized system state • A cloud reconfiguration algorithm is developed to transfer the cloud from the current state to the optimized one

Thank you