Hyper-converged cluster (CEPH design) document

[Axl_fzetafs]

I know it's a bit OT since th question pertains one of the aspects of ProxMox, but is there any dcoument that may explain a bit what ar the considerations/points to take into account when building up a hyper-converged cluster with ProxMox using CEPH? For instance, may I have nodes with different storage capacity? Is the solution prrof to the sudden unavailability (electricity cut) of a node?
The whole goal is for personal usage.

TIA,

Alex

 

[tom]

A good start:

https://forum.proxmox.com/threads/proxmox-ve-ceph-benchmark-2020-09-hyper-converged-with-nvme.76516/

 

[pille99]

A good start:

https://forum.proxmox.com/threads/proxmox-ve-ceph-benchmark-2020-09-hyper-converged-with-nvme.76516/

its pretty old and some keypoints are depleted at 2023. will there be an updated version any time soon ?
i just found one time a pretty good documentation and walkthru about performance and performance improvements and configuration from an university but never found it again.

 

[tom]

its pretty old and some keypoints are depleted at 2023. will there be an updated version any time soon ?
i just found one time a pretty good documentation and walkthru about performance and performance improvements and configuration from an university but never found it again.

Which key points do you talk about? The executive summary is for sure still valid in 2023.

If you have ideas to improve, share it here.

And yes, new benchmark hardware is ordered/planned.

 

[alexskysilk]

its pretty old and some keypoints are depleted at 2023.

Really not much changed- not until crimson comes out.

 

[Axl_fzetafs]

Really not much changed- not until crimson comes out.

What's Crimson?

 

[tom]

https://ceph.io/en/news/blog/2023/crimson-multi-core-scalability/