Hello,
My Cloudstack journey as a sysadmin and (ex)cloud operator started with
Openstack. I've since been through OpenNebula, oVirt etc.
I'm obviously wearing the Cloudstack hat now, but that's because it hits
the sweet spot between features and complexity, especially if you're
with a small ops team.
I don't see anything in the general HPC requirements that Cloudstack
cannot satisfy, same for your points below.
You'll need:
- Scalability - checked
- Flexibility - checked
- Cost effectiveness - dooh
- Resource management - checked
- Multi-tenancy - checked
- Interoperability and automation - checked (open, fully documented api,
automation tools available - terraform, ansible etc)
- Hearty community support - checked
- Commercial support, should you need it - checked
As far as k8s is concerned, I think CAPI/CAPC can be an answer.
On Kubeflow, should you wish to use it - knowing nothing about it other
that it's ML focused - you will need to see what are the GPU
requirements there, if any, that could be a pain point, albeit surely
addressable one way or another.
Taking my Cloudstack hat off, in an ideal world you would take some time
to test the 3 solutions thoroughly (cloudstack, openstack, opennebula)
and see which matches your needs and capabilities best and also have a
go at maintainability, because getting off the ground is the easy bit.
And lastly, if you have paid attention to this project, you will have
noticed that it's a living, breathing one, if something is wrong or
missing today, it will be fixed tomorrow, metaphorically speaking.
Looking at the roadmap of the past few years, it's just crazy how many
new awesome things materialised.
HTH
On 2023-04-21 00:13, Axel Baudot wrote:
Hello dear CloudStack users,
As a team sitting at a crossroad to chose a solution to provision
ressources for a HPC oriented datacenter: can a strong case be made for
the use of Cloudstack over Openstack, OpenNebula or Kubernetes?
In particular I have the following concerns :
1. The company provides infrastructure to clients, as well as full
blown project development, so it trying to set up both a public cloud
with resource billing and an internal development platform. This is the
use case covered by CloudStack, right? Or could the flexibility of
OpenStack be needed?
2. It is also heavily geared toward HPC. I have seen little mention of
HPC with CloudStack, while it is heavily advertised in OpenStack world,
due to the ties with CERN and such. Can you think of limitations of
CloudStack for this use case? I am also interested in case studies or
any reading material on this combination.
3. The team could standardize on containers soon (especially for dev
environments). I would tend to do a simple VM + Docker workflow. Is it
standard?
4. I am all for simplicity of operation and maintenance. I think
CloudStack could really shine here, right? I also think that Kubernetes
should be avoided because of this. Thing is, Kubeflow is getting some
attention and we might have to support it... Is CKS considered mature
for production and a viable solution in this case?
Thanks for reading, any food for thoughts will be very much
appreciated.
Best,
Axel
