Broadcast latencies in the Filecoin network

[guy-goren]

Hi,

We're working on analyzing the security of Filecoin's Consensus mechanism, which significantly relies on timing assumptions. To define a model that best captures the reality it would be extremely helpful to know the current latencies in Filecoin's mainnet. In particular, the latencies associated with broadcasting. Any information would be valuable! (Mean latency per sender/receiver, complete distribution of latencies, 95th percentile, etc.)

cc @yiannisbot @sa8 @jsoares

[yiannisbot]

Thanks for raising this - important topic indeed! Just a quick note to say that we've already started planning on this. It has been on our plan to carry out this study for a while now, but got de-prioritised over other items.

@cortze and @leobago will be looking into this in the coming month or so. We'll ping you when we get started.

[jsoares]

Thanks! In the meantime, we've procured some historical data from sentinel (https://github.com/protocol/pldw/issues/146) but we're still trying to figure out how good it is -- and, due to the passive collection and no trusted source timing, it comes with certain structural limitations. Looking forward to working with you!

[leobago]

Thanks @jsoares, for some reason the link is not working for me.

[jsoares]

Any chance you already have a timeline/estimate for the study? Trying to make sure I align our roadmap 🙏

[yiannisbot]

@cortze do you have an estimate of when you'd be starting this study? I guess not before the end of October, given the paper deadline you have, but would be great to schedule it straight after, WDYT?

[cortze]

Apologies for coming late to the thread @jsoares , It will be my next priority after the submission of a paper that we are working on. So I will be able to start working on it after the IPFS Camp 🚀