Implement storage QoS using OpenStack?

asked 2016-01-20 20:49:20 -0500

racampbe gravatar image

Hi all,

This is a question primarily oriented towards OpenStack developers, and may be a little vague or naive. We are a team of undergrad students working on creating a layer of storage QoS in a SAN or cloud environment. For instance, payroll processing storage access above client home directory access. We are intending to modify the Linux kernel and have some good ideas for implementation - but we are very interested in OpenStack and if possible contributing something valuable in this platform if possible.

We have a very limited understanding of OpenStack and its overall architecture/purpose. The purpose of this question is to get feedback on the feasibility of using OpenStack before we really dive in. Our current, most likely candidate for how this will be done is using an out-of-band management server with a thin layer of understanding of storage performance to coordinate/throttle clients.

Is this a potential use case of OpenStack? Does it fit in with its purpose? Is there a better way of doing this using OpenStack ideas/protocols that already exist?

We appreciate any feedback and insight that you could provide.

Thank you!

edit retag flag offensive close merge delete