180 likes | 287 Views
COMMA: Coordinating the Migration of Multi-tier applications. Presenter: Zhaolei (Fred) Liu. Jie Zheng* T.S Eugene Ng* Kunwadee Sripanidkulchai† Zhaolei Liu* *Rice University, USA †NECTEC, Thailand. Live migration in cloud. For users Provision services
E N D
COMMA: Coordinating the Migration of Multi-tier applications Presenter: Zhaolei (Fred) Liu Jie Zheng* T.S Eugene Ng*Kunwadee Sripanidkulchai† Zhaolei Liu* *Rice University, USA †NECTEC, Thailand
Live migration in cloud • For users • Provision services • Price and performance • For cloud providers • Hardware maintenance • Resource relocation
Multi-tier application in cloud Presentation tier • In cloud, a multi-tier application runs on multiple VMs • The VMs hosting a multi-tier application need to communicate with each other Logic tier Data tier
How to migrate a group of VMs? • Sequential migration: migrate VMs one by one
How to migrate a group of VMs? • Parallel migration: migrate all VMs at the same time
Problem with sequential migration • Application performance degradation caused by component split during migration
Problem with parallel migration • VMs still may not finish migration at the same time • Static factors: VM disk size, memory size, etc. • Dynamic factors: network bandwidth, application workload, etc.
COMMA: Coordinating the Migration ofMulti-tier Applications 1s • Formulation & objective • System design • Algorithms • Implementation & evaluation
Formulation & Objective • Minimizing the migration’s impact on application performance • Define impact as the time when VMs are split • Define impact as the volume of traffic impacted by migration • TM: traffic matrix • ti: the migration finish time of the ith VM Not ideal! Minimize
COMMA: Periodic adaptation and coordination Controller • Hypervisor Messages: • Migration progress • Available bandwidth • Disk dirty rate and • memory dirty rate • (Pacer*) • Controller Messages: • Start migration • Set migration speed Hypervisor Hypervisor Hypervisor * J. Zheng, T. S. E. Ng, K. Sripanidkulchai, and Z. Liu. Pacer: A progress management system for live virtual machine mi- gration in cloud computing. IEEE Transactions on Network and Service Management, 10(4):369–382, Dec 2013.
Coordination in the first stage of pre-copy • Coordinate pre-copy of all VMs to finish at the same time Stage 2 Stage 1 vm1 30 80 VM1 Pre-copy vm2 vm3 VM2 Pre-copy VM3 Pre-copy 20 50 VM4 Pre-copy vm4 Time Migration End Migration Start Communication Graph (KBps)
Inter-group scheduling in the second stage of dirty iteration and memory migration • Meet the bandwidth limit by dividing VMs into different groups Stage 2 Stage 1 vm1 30 80 VM3 VM4 VM1 VM2 VM1 Pre-copy vm2 vm3 VM2 Pre-copy VM3 Pre-copy 20 50 VM4 Pre-copy vm4 Time Migration End Migration Start Communication Graph (KBps) 20MBps 10MBps vm3 vm1 30MBps 5MBps vm4 20MBps vm2 Maximal dirty rate Available bandwidth
Intra-group scheduling • Maximize bandwidth utilization and minimize performance degradation by scheduling dirty iteration inside each group in the second stage VM1 VM3 VM2 VM1 VM3 VM2 VM1 Bandwidth Bandwidth Bandwidth VM2 VM3 Time Time Time Performance Degradation;Short Migration Time Start at the same time; Finish at different time Start at the same time; Finish at the same time No Performance Degradation;Long Migration Time No Performance Degradation; Short Migration Time Start at different time; Finish at the same time
Implementation & Evaluation: • Fully implemented COMMA on KVM platform, QEMU version 0.12.50 • Used SpecWeb and RUBiS as the application • Fully Evaluated COMMA on various scenarios
COMMA is able to reduce application performance degradation Migrating 3-VM RUBiS using COMMA Average response latency (ms) Time(s)
Compared to COMMA, sequential migration incurs high application performance degradation Migrating 3-VM RUBiS using sequential migration Average response latency (ms) Time(s)
COMMA is able to minimize migration impact 3.6 MB More results: vary the VM number, placement, workload, and migration to EC2
Summary &Advantages • COMMA is able to minimize application performance degradation during migration • COMMA has tiny overhead • Efficient heuristic algorithms • Computation time less than 10 ms • COMMA is application independent • Run-time adaptation • All measurements are on hypervisor level • COMMA has great applicability • Designed for pre-copy model (KVM, XEN) • Easily adapt to other migration models