What
The dashboard today is minerId (SP instance in Filecoin protocol) focused. In reality, we know that SP operations (whether individuals, entities, etc.) are often composed of multiple minerIds. Ideally in the data model we'd have this relationship so it would be possible to bring up the performance an "SP entity/operation", looking at an aggregate view across multiple related minerIds.
Why
This came up in https://protocollabs.notion.site/2024-12-03-Client-Success-Meeting-150837df73d4801eb9cae595e2ea1d7c?pvs=4 . It's relevant from a market perspective given onramps/aggregators care about making deals with an actual "SP entity/operation" rather than a specific minerId.
How
I (@BigLep) don't know the way to accomplish this. There is no mapping on chain of minerId to peerId. Maybe we can look at the origin address of WindowPoSt messages and aggregate that way? (I assume that is still an approximation, but likely gets decently close.). I also assume other thought has been given to this in the past.
Other Notes
- Maybe this belongs in https://github.com/davidgasquez/filecoin-data-portal and this issue just become about rendering that relationship?
What
The dashboard today is minerId (SP instance in Filecoin protocol) focused. In reality, we know that SP operations (whether individuals, entities, etc.) are often composed of multiple minerIds. Ideally in the data model we'd have this relationship so it would be possible to bring up the performance an "SP entity/operation", looking at an aggregate view across multiple related minerIds.
Why
This came up in https://protocollabs.notion.site/2024-12-03-Client-Success-Meeting-150837df73d4801eb9cae595e2ea1d7c?pvs=4 . It's relevant from a market perspective given onramps/aggregators care about making deals with an actual "SP entity/operation" rather than a specific minerId.
How
I (@BigLep) don't know the way to accomplish this. There is no mapping on chain of
minerIdtopeerId. Maybe we can look at the origin address of WindowPoSt messages and aggregate that way? (I assume that is still an approximation, but likely gets decently close.). I also assume other thought has been given to this in the past.Other Notes