Ask Your Question
4

Can i scale up/scale down computing resource(i.e. cpu, ram, disk) on running vm instances without shutting it down?

asked 2013-05-31 07:27:11 -0600

sudhir gravatar image

updated 2013-06-08 22:32:37 -0600

fifieldt gravatar image

Can i add more computing power i.e. cpu, ram, disk to live/running vm instances without shutting it down?

can i set a threshold in such a way that if cpu usage exceeds 80% then automatically add additional computing power such as add couple of more CPU cores to running VM instance, can do this ? how do i configure this in nova? pls suggest.

Thanks in advance.

edit retag flag offensive close merge delete
add a comment

2 answers

Sort by » oldest newest most voted
5

answered 2013-05-31 08:32:34 -0600

Sagar Ratnakara Nikam gravatar image

You can use the resize API to add a different flavor to that VM. Check this doc

http://docs.openstack.org/api/openstack-compute/2/content/Resize_Server-d1e3707.html

This is a user initiated action. You cannot set a threshold and resize it automatically on reaching that threshold.

To set threshold and the increase the size of the VM, you can do the following

HealthNMon, a opensource “Cloud Resource Monitor” provides among other features, a feature for fetching CPU utilization of a VM launched in Openstack

https://github.com/stackforge/healthnmon

You can write a custom script/cron job, which keeps checking the CPU utilization using HealthNMon, and when it reaches the threshold, call the resize API and change the flavor

edit flag offensive delete link more
add a comment
2

answered 2013-05-31 07:37:50 -0600

Alen Komljen gravatar image

Short answer no, there is no such option in OpenStack or at any public cloud provider as I know. On running instances you can just edit security group, attach volumes or associate floating IP-s. So to change memory or cpu, you can just create snapshot from running instance and spin up new one from that snapshot using different size.

edit flag offensive delete link more
add a comment

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

subscribe to rss feed

Stats

Asked: 2013-05-31 07:27:11 -0600

Seen: 14,031 times

Last updated: Jan 20 '14

Related questions

trigger heat autoscale group by an other resource

Ceilometer Alarm insufficient Data

Sending Heat Scale URLs to Instances

IncompatibleOjectVersion [closed]

Nova-compute does not show up in service list

can't create bridge with the same name - Havana [closed]

where to install cinder

Unable to create instances icehouse due to rabbit mq timeout

br-int internal vlan tag

how can i add a second compute node ?