You are on page 1of 16

1

A Performance Study on the VM Startup Time in the Cloud


Cloud 2012
(June 25, Honolulu, Hawaii)

Ming Mao, Marty Humphrey CS Department, University of Virginia

Introduction
2

One great advantage of the cloud is dynamic scalability, BUT


Only

meaningful when the compute resource is available in time The fact is that acquired instances are NOT ready immediately

Why study the cloud VM startup time


Important

to time sensitive applications Important to resource auto-scaling mechanisms Inconsistent definitions across cloud providers To set up a reference for the community To know the difference from two years ago

Experiment setup
3

Type
m1.small m1.small Small Small Small Type IV Type IV

OS Image Amazon EC2


Linux(Fedora) ami-48aa4921 Windows (Win Server 2008) ami-fbf93092

Location
us-east-1a us-east-1a South Central US South Central US South Central US N/A N/A

AWS, Windows Azure, Rackspace Oct 15th, 2011 Feb 15th 2012 A client periodically collects the instance startup information By different factors time, image size, machine type, location, etc.

Windows Azure
WebRole default WebRole app in Azure SDK WorkerRole default WorkerRole app in Azure SDK VMRole Win Server 2008R2

Rackspace
Linux (Fedora) flavor 71 Windows (Win Server 2008R2) flavor 28

VM
Linux VMs Windows VMs WebRoles WorkerRoles

Definition of VM startup time


The first successful ssh login The first successful remote desktop connection The first successful http request The first successful logging

Results (1) by time


4

Cloud
EC2 EC2 Azure Azure Azure Rackspace Rackspace

OS
Linux Windows WebRole WorkerRole VMRole Linux Windows

Average VM startup time


96.9 seconds 810.2 seconds 374.8 seconds 406.2 seconds 356.6 seconds 44.2 seconds 429.2 seconds

Results (2) by image size


5

Cloud EC2 Azure Rackspace

Average Data Transfer Rate between VM and Image Store 10.9 MB/s 1.1MB/s 22.5 MB/s

Results (3) by machine type


6

Azure, Rackspace the VM startup time goes longer as the instance type goes larger EC2 no significant differences among different instance types, except the micro ones

Results (4) by location


7

No significant differences among locations EC2 - The newly established data center has a slightly slower startup performance and greater variance

Results (5) multiple instances


8

EC2 VM startup time is relatively constant. Azure the last VM instance sometimes took significantly longer than the first one. RS does not support instance acquisition in batch

Results (6 - 1) spot instances, by time


9

Spot instances have a longer waiting time but similar VM startup performance compared to on-demand instances

Results (6 - 2) spot instances, by vm type


10

Similar to on-demand instances, the VM startup time shows no significant differences across different machine types

Results (6 - 3) spot instances, by location


11

The US east region shows longer spot instance acquisition time

Results (6 - 4) spot instances, multiple instances


12

Similar to on-demand instances, spot instances show constant performance across all the 16 machines acquired at the same time

Results (6 - 5) spot instances, by real-time price


13

No significant relations between the VM startup time and the real-time price

Result (7) other facts


14

Spot instances are not always cheaper than ondemand VMs Acquisition requests are not always successfully served (unusable machines).

Cloud
EC2 Azure Rackspace

Success Rate
99.2% 99.6% 92.0%

VM Release Time
3-8 seconds 8-21 seconds 3-8 seconds

Summary
15

Summary

VM startup time is independent of time of the day. EC2, Rackspace - Windows instances take 9 times longer than Linux instances. Azure - all three Role instances show similar performances. Both the size of the OS image and the instance type can largely affect the VM startup time. spot instances - longer VM startup time and greater variance RS had a higher instance acquisition failure rate (8%) > EC2 (0.8%) & Azure (0.4%). Azure 200-second improvement and smaller variances; EC2 performance does not change.

Questions
16

Thank you!

You might also like