[Openstack-operators] Question for deployers: how much variation is there in hardware in your deployments
Tim.Bell at cern.ch
Sun Dec 1 19:37:59 UTC 2013
- We're currently running around 10 different hardware configurations
- Our hardware configurations are based on the requirements of public tendering, i.e. cheapest compliant bid with assembly in one of CERN's member states. The basic criteria is the total cost of ownership (purchase + electricity + maintenance) / performance based on a subset of HEP Spec 2006.
- The configurations vary year by year. Recently, Supermicro twins and twin**2 configurations are coming out on top.
- We provide our users with the Amazon flavors (m1.*) and a specific set of flavors for High Energy Physics based on 2GB/core. Additional flavors can be requested but are discouraged (see http://openstack-in-production.blogspot.fr/2013/08/flavors-english-perspective.html for details)
- We do not currently delegate flavors for project administration. We want to review why the standard flavors are not sufficient (and the limitations in OpenStack cells support means we can't delegate it on Grizzly)
- We do try to find some standard flavors to ensure that the packing of VMs on physical hardware is reasonable. A VM needing 1 core and 48 GB of memory will clearly leave idle cores around.
> -----Original Message-----
> From: Robert Collins [mailto:robertc at robertcollins.net]
> Sent: 01 December 2013 20:17
> To: openstack-operators at lists.openstack.org
> Subject: [Openstack-operators] Question for deployers: how much variation is there in hardware in your deployments
> This is for input to the TripleO design prioritisation discussions we're having at the moment.
> We have been assuming that everyone has almost no variation on hypervisors, but specialised hardware configuration for control plane /
> data storage / compute - so typically 3 profiles of machine.
> But, we'd like a little data on that.
> So - if you're willing, please reply (here or directly to me) with answers to the following q's:
> - how many hardware configurations do you have
> - are they deliberate, or organic (e.g. optimising for work vs your vendor couldn't deliver the same config anymore)
> - if you have multiple hypervisor hardware configurations, do you expose that to your users (e.g. using different flavors to let users choose
> which servers to run vms on) - a story for that might be high performance computing vs densely packed vms.
> Robert Collins <rbtcollins at hp.com>
> Distinguished Technologist
> HP Converged Cloud
> OpenStack-operators mailing list
> OpenStack-operators at lists.openstack.org
More information about the OpenStack-operators