News for hackers. Stuff that matters.
Re:Metal as a Service by Maakuth on Tuesday June 03, 2014 @10:01AM (#7840199)
This is an interesting concept and a good addition to all the IaaS offerings around. Just one year ago, while doing my Master's thesis, I pondered the possiblity of creating an IaaS that didn't involve virtual machines but physical servers instead. This should be great especially for high performance and special hardware needs. Next step would be all sorts of mixed blends between virtual and physical servers: real CPUs and GPUs, but virtualized storage, networking or what have you.
Re:Metal as a Service by rwmj on Tuesday June 03, 2014 @10:12AM (#7840270)
Is this so novel? Both Amazon and OpenStack can provision and manage baremetal instances.
Re:Metal as a Service by pquerna on Tuesday June 03, 2014 @11:31AM (#7840782)
Amazon does not have a 'baremetal' instance type.

OpenStack 'can', but in practice the baremetal nova driver[0] has many issues, and the newer OpenStack Ironic Project[1] is not yet ready for production workloads.

Additionally, both baremetal and OpenStack Ironic have a focus on 'triple-O', or OpenStack on OpenStack[2] use cases -- that is using these projects to bootstrap an 'undercloud', for the hardware to be used by a higher level virtualized cloud.

[0] - https://wiki.openstack.org/wiki/Baremetal

[1] - https://wiki.openstack.org/wiki/Ironic

[2] - https://wiki.openstack.org/wiki/TripleO

Re:Metal as a Service by giulivo on Tuesday June 03, 2014 @03:45PM (#7842315)
> Additionally, both baremetal and OpenStack Ironic have a focus on 'triple-O', or OpenStack on OpenStack[2] use cases

Well, I'm sure they have issues but what gave you the impression they are focused on TripleO use cases?

As a TripleO dev I just use them.

Compared to more 'standalone' softwares like MAAS, these tools are integrated already into OpenStack, allowing you to deploy images on either virtual instances or baremetal without adding more software layers in charge of the images hosting, authentication, dhcp management and what not.

Re:Metal as a Service by rwmj on Tuesday June 03, 2014 @11:34AM (#7840805)
Fair enough.

I was totally convinced that Amazon had baremetal instances available, but now I cannot find anything, so that could probably be a figment of my imagination.

Re:Metal as a Service by kikoreis on Tuesday June 03, 2014 @12:30PM (#7841163)
All of Amazon's EC2 compute instances are run on top of Xen; there is an interesting Reddit AMA post with a lot of detail: http://www.reddit.com/r/IAmA/comments/1e5o4p/iaman_exaws_eng...
Re:Metal as a Service by rwmj on Wednesday June 04, 2014 @04:27AM (#7844578)
That is a very interesting thread, thanks. I think I was confusing dedicated tenancy with baremetal. Dedicated tenancy of course still runs on Xen.
Re:Metal as a Service by YokoZar on Tuesday June 03, 2014 @05:10PM (#7842742)
I believe at one point the GPU instances were being given direct hardware access to the GPU as virtualized GPUs weren't yet a thing. So the GPU part was sort of baremetal I guess.
Re:Metal as a Service by wmf on Tuesday June 03, 2014 @05:16PM (#7842779)
GPU instances are still Xen VMs, not bare metal.
Re:Metal as a Service by teepo on Tuesday June 03, 2014 @12:29PM (#7841154)
SoftLayer has been doing this since 2005. One API to provision and deploy compute on demand physical or virtual. Over 90% of their customer bases uses bare metal servers.
Re:Metal as a Service by marcoceppi on Tuesday June 03, 2014 @10:04AM (#7840223)
MAAS can already manage virtual machines to an extent (through libvirt) but adding storage support would be awesome.
Re:Metal as a Service by angersock on Tuesday June 03, 2014 @11:31AM (#7840785)
You've reinvented what we used to call dedicated servers. There is nothing new under the sun.

Might as well call it physicalization. :)

Re:Metal as a Service by tealeg on Wednesday June 04, 2014 @04:47AM (#7844621)
:-) Of course it's not inventing anything, it's just providing a convenient way to commission those servers that can also be driven by other tools. If you bring a lot of server resource on line this is very handy. Juju, for example, can put things on MAAS servers, and one of the things it could put on there is a virtualised environment that could also be managed by MAAS and juju to deploy other workloads. MAAS knows some stuff about the hardware too, for example, how many network cards the machine has and on which networks, which means you can interrogate MAAS to find machines suitable for a given workload.