Fast Server Deprovisioning through Scatter-Gather Live Migration of Virtual Machines

Deshpande, Umesh; You, Yang; Chan, Danny; Bila, Nilton; Gopalan, Kartik

doi:10.1109/cloud.2014.58

Cited by 25 publications

(10 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using cloudsim, our proposed algorithm managed to move the extra tasks from overloaded VMs to more powerful VM, aiming to complete the process with minimized timespan. According to the output results, tasks (0, 1,2,3,4,5,6,7,8,9) were allocated to VMs (4,5,3,3,1,1,3,5,3,4) which is shown in Figure.4.As Walker [17] noted, the minimum time needed Table 1-Pseudo code of Endocrine-PSO based algorithm.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…They showed that dynamic allocation can improve the server performance while enhancing the quality of service. Deshpande [8]offered a new method called scatter-Gather which is minimizing the eviction time by applying the pre/post -copy method. In this method they tryied to decouple the source and the destination by scattering the VM's memory state to midlle boxes.…”

Section: Related Work In Load Balancing Optimizationmentioning

confidence: 99%

See 1 more Smart Citation

Load balancing optimization in cloud computing: Applying Endocrine-particale swarm optimization

Aslanzadeh

Chaczko

2015

2015 IEEE International Conference on Electro/Information Technology (EIT)

View full text Add to dashboard Cite

Load balancing optimization is categorized as NP-hard problem, playing an important role in enhancing the cloud utilization. Different methods have been proposed for achieving the system load balancing in cloud environment. VM migration is one of these techniques, proposed to improve the VMs' functionality. Despite of the advantageous of VM migration, there are still some drawbacks which urged researchers to improve VM migration methods. In this paper we propose a new load balancing technique, using Endocrine algorithm which is inspired from regulation behavior of human's hormone system. Our proposed algorithm achieves system load balancing by applying selforganizing method between overloaded VMs. This technique is structured based on communications between VMs. It helps the overloaded VMs to transfer their extra tasks to another under-loaded VM by applying the enhanced feed backing approach using Particle Swarm Optimization (PSO). To evaluate our proposed algorithm, we expanded the cloud simulation tool (Cloudsim) which is developed by University of Melbourne. The simulation result proves that our proposed load balancing approach significantly decreases the timespan compared to traditional load balancing techniques. Moreover it increases the Quality Of Service (QOS) as it minimizes the VMs' downtime.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Related Work In Load Balancing Optimizationmentioning

confidence: 99%

Load balancing optimization in cloud computing: Applying Endocrine-particale swarm optimization

Aslanzadeh

Chaczko

2015

2015 IEEE International Conference on Electro/Information Technology (EIT)

View full text Add to dashboard Cite

show abstract

“…It is used as a default migration technique by Xen [2], VMware [20], and KVM [25] hypervisors. Whereas, post-copy is known for its low network overhead and allows quick consolidation [17] or eviction [13] of VMs.…”

Section: Introductionmentioning

confidence: 99%

Traffic-Sensitive Live Migration of Virtual Machines

Deshpande

Keahey

2015

2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

Self Cite

View full text Add to dashboard Cite

In this paper we address the problem of network contention between the migration traffic and the Virtual Machine (VM) application traffic for the live migration of co-located Virtual Machines. When VMs are migrated with pre-copy, they run at the source host during the migration. Therefore the VM applications with predominantly outbound traffic contend with the outgoing migration traffic at the source host. Similarly, during post-copy migration, the VMs run at the destination host. Therefore the VM applications with predominantly inbound traffic contend with the incoming migration traffic at the destination host. Such contention increases the total migration time of the VMs and degrades the performance of the VM application. Here, we propose a traffic-sensitive live VM migration technique to reduce the contention of migration traffic with the VM application traffic. It uses a combination of pre-copy and post-copy techniques for the migration of the co-located VMs (those located on the same source host), instead of relying on any single pre-determined technique for the migration of all the VMs. We base the selection of migration techniques on the VMs' network traffic profiles so that the direction of migration traffic complements the direction of the most VM application traffic. We have implemented a prototype of traffic-sensitive migration on the KVM/QEMU platform. In the evaluation, we compare trafficsensitive migration against the approaches that use only pre-copy or only post-copy for VM migration. We show that our approach minimizes the network contention for migration, thus reducing the total migration time and the application degradation. 2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing 978-1-4799-8006-2/15 $31.00

show abstract

“…VM migration enables the movement from an online server to a new location, allowing an efficient usage and allocation of resources on-demand, ensuring the accomplishment of SLAs and minimizing costs. Some proposals have tried to mitigate these issues by providing algorithms [Kashyap et al, 2014, Lu et al, 2015, and new metrics [Deshpande et al, 2014] for the evaluation, optimization and scheduling of VM migration.…”

Section: Related Workmentioning

confidence: 99%

Component-wise Application Migration in Bidimensional Cross-cloud Environments

Carrasco

Durán

Pimentel

2017

Proceedings of the 7th International Conference on Cloud Computing and Services Science

View full text Add to dashboard Cite

Abstract:We propose an algorithm for the migration of cloud applications' components between different providers, possibly changing their service level between IaaS and PaaS. Our solution relies on three of the key ingredients of the trans-cloud approach: a unified API, agnostic topology descriptions, and mechanisms for the independent specification of providers. We show how our approach allows us to overcome some of the current interoperability and portability issues of cloud environments to propose a solution for migration, present an implementation of our proposed solution, and illustrate it with a case study and experimental results.

show abstract

Fast Server Deprovisioning through Scatter-Gather Live Migration of Virtual Machines

Cited by 25 publications

References 16 publications

Load balancing optimization in cloud computing: Applying Endocrine-particale swarm optimization

Load balancing optimization in cloud computing: Applying Endocrine-particale swarm optimization

Traffic-Sensitive Live Migration of Virtual Machines

Component-wise Application Migration in Bidimensional Cross-cloud Environments

Contact Info

Product

Resources

About