Lab Memory usage!

[sdargoeuves]

I've never spent any time investigating this, our lab had about 28x cEOS (amongst other vendors), and the cEOS are fast to boot, that's all I needed.
Now that we can show the memory usage of the nodes in the lab (netlab status --memory) I've actually realised how much RAM the cEOS are using 😱 each cEOS takes an average of 1.7GB (ranging between 1.6–2.0GB in practice), that's becoming quite a heavy memory user.

Being made aware of how much memory each node is eating, I decided to use different vendors where convenient... and now instead of those 28 Arista devices, I have kept 12, and converted the rest to a mix of different nodes (plus added a 3rd spine in the CLOS fabric, because why not!)

Here is what the math looks like:

• Before (All cEOS):
  28× cEOS @ ~1.7 GB = ~47.6 GB of RAM

• After (mix):
  12× cEOS (~1.7 GB) = ~20.4 GB
  2× FRR (~35 MB) = ~70 MB
  12× IOL-L2 (~165–320 MB) = ~2.2 GB
  3× IOL (~190 MB) = ~570 MB
  Total: ~23.2 GB of RAM

The memory footprint was cut in half, saving over 24 GB of RAM.

Takeaway: Just because it's easy to use cEOS devices, it may not always be the right choice. I had no idea the IOL and IOL-L2 images were so much lighter, and took for granted cEOS as my default. I'm glad things have changed, I've now got more room to add new stuff to the lab!

Now if someone has a good idea on how to reduce the memory consumption of the Cisco SDWAN Manager... I'm all ears 😄

╰─❯ netlab status --memory     
  status:      started
  topology:    topology.yml
  provider(s): libvirt,clab
  memory used: 134.060GiB

┏━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━┳━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃ node             ┃ device   ┃ image                                      ┃ mgmt IP       ┃ connection  ┃ provider ┃ VM/container               ┃ status               ┃ memory    ┃
┡━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━╇━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ sdwan-manager    │ linux    │ vrnetlab/cisco_sdwan-manager:20.16.1       │ 10.194.56.130 │ docker      │ clab     │ clab-ml-1-sdwan-manager    │ Up 7 hours (healthy) │ 32.620GiB │
├──────────────────┼──────────┼────────────────────────────────────────────┼───────────────┼─────────────┼──────────┼────────────────────────────┼──────────────────────┼───────────┤
│ sdwan-controller │ linux    │ vrnetlab/cisco_sdwan-controller:20.16.1    │ 10.194.56.131 │ docker      │ clab     │ clab-ml-1-sdwan-controller │ Up 7 hours (healthy) │ 1.013GiB  │
├──────────────────┼──────────┼────────────────────────────────────────────┼───────────────┼─────────────┼──────────┼────────────────────────────┼──────────────────────┼───────────┤
│ sdwan-validator  │ linux    │ vrnetlab/cisco_sdwan-validator:20.16.1     │ 10.194.56.132 │ docker      │ clab     │ clab-ml-1-sdwan-validator  │ Up 7 hours (healthy) │ 884.3MiB  │
├──────────────────┼──────────┼────────────────────────────────────────────┼───────────────┼─────────────┼──────────┼────────────────────────────┼──────────────────────┼───────────┤
│ edge1x01         │ cat8000v │ vrnetlab/cisco_c8000v:controller-17.15.04c │ 10.194.56.133 │ network_cli │ clab     │ clab-ml-1-edge1x01         │ Up 7 hours (healthy) │ 4.034GiB  │
├──────────────────┼──────────┼────────────────────────────────────────────┼───────────────┼─────────────┼──────────┼────────────────────────────┼──────────────────────┼───────────┤
│ edge2x01         │ cat8000v │ vrnetlab/cisco_c8000v:controller-17.15.04c │ 10.194.56.134 │ network_cli │ clab     │ clab-ml-1-edge2x01         │ Up 7 hours (healthy) │ 4.032GiB  │
├──────────────────┼──────────┼────────────────────────────────────────────┼───────────────┼─────────────┼──────────┼────────────────────────────┼──────────────────────┼───────────┤
│ edge3x01         │ cat8000v │ vrnetlab/cisco_c8000v:controller-17.15.04c │ 10.194.56.135 │ network_cli │ clab     │ clab-ml-1-edge3x01         │ Up 7 hours (healthy) │ 4.120GiB  │
├──────────────────┼──────────┼────────────────────────────────────────────┼───────────────┼─────────────┼──────────┼────────────────────────────┼──────────────────────┼───────────┤

[sdargoeuves]

Thank you @ipspace and all the contributors for this amazing project!

[ipspace]

Thank you @ipspace and all the contributors for this amazing project!

... and thank you for all the cool contributions, like the "display memory usage" thingy ;))