Eclair architecture

[kiwiidb]

On a high level, how does the architecture of Eclair make it possible to run a single node in a distributed way, and especially what are the differences with LND's (or CLN's) architecture where it seems that the only good option of doing HA is by using a phantom node + multiple, seperate real nodes?

[t-bast]

I recommend having a look at our cluster documentation and architecture blog post, which should contain details about eclair's setup. If you have questions about specific details of this setup, don't hesitate to add them and we'll dive more deeply into it.