Ask Your Question

Trove is a database as a service solution or a virtual machine image database?

asked 2014-05-03 08:00:36 -0500

mr_aliagha gravatar image


I am a new user to openstack trove. As far as I found (from the process of creating datastore and database in trove) trove works like this: For each datastore instance there is a nova-compute image that this instance will be launch on that (and also a cinder storage assigned to this instance) Therefore there is not a centralized database which could be manage by openstack administrator. As far as I know there are two types of cloud database: 1)virtual machine image database and 2)dbaas. For dbaas it should not be like having a virtual machine instance per each database and database provisioning should be manageable by system administrator (not client). Therefore could somebody explain for me how trove works and how could we consider it dbaas and not virtual machine image database?

edit retag flag offensive close merge delete


Not sure, but from that what I could read it seems you can create different database servers for different tenants and create your mysql, redis, mongo, etc. dbs in each instance with backup and recovery support and I guess trove uses heat templates to provide real DBaaS.

cloudssky gravatar imagecloudssky ( 2014-05-03 09:40:37 -0500 )edit

2 answers

Sort by ยป oldest newest most voted

answered 2014-05-03 14:02:47 -0500

Trove is a DBaaS solution that integrates with OpenStack. You could find a first introduction and additional info on Trove here.

The OpenStack image service is Glance. Here its documentation. [It basically stores images that could be used as references to spawn VMs on your infrastructure]

edit flag offensive delete link more


please take a look at (this), Are you still think that trove is a database as a service solution?

mr_aliagha gravatar imagemr_aliagha ( 2014-05-03 16:20:24 -0500 )edit

answered 2016-10-18 02:31:22 -0500

Ajaykanth gravatar image

So can we have trove working without having a nova instance. Lets say I install keystone and Trove on one VM. To use postgres, will install the corresponding agent. Will trove be able to work in this scenario?

edit flag offensive delete link more


No, I do not think so. You should have a minimal Openstack Setup (Keystone, Nova, Neutron (or at least Nova Network), Glance and Trove). Try reading this.

Antonio G. gravatar imageAntonio G. ( 2016-10-19 04:50:15 -0500 )edit

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

1 follower


Asked: 2014-05-03 08:00:36 -0500

Seen: 370 times

Last updated: May 03 '14