Feature request: explorer | subgraphs | characteristics about the indexing process

[WayneWayner]

I think it would be very informative if there were some additional information for each subgraph like :

• How many blocks the subgraph currently has
• What is the effort to index the subgraph (time)
• What is the approximate impact on the database
• How complex are querys

This information could be collected continuously from results of previous indexers. For example, it was already known that Moloch would sync faster than Uniswap.

[crypt1d]

+1

Another important piece of information would be what eth node capabilities are required for indexing (full/archives/traces).

[davekaj]

Good points. Some of this would require engineering work.

• How many blocks the subgraph currently has - not sure exactly what you mean, but if it is on mainnet you then it would have the same amount of blocks as mainnets head.
• Thats a great point. I think it's unlikely we will have anything like this for mainnet. As far as I know, there is no way to know ahead of time. It would just come from experience. We could for sure make an effort to coordinate out of protocol to do this at the start of the network, and I guess this is what will happen organically.
• Also tough, seems like it will have to happen organically through indexer chat
• Also tough, seems like it would have to happen organically through indexer chat

These are really good suggestions but I think if I am being realistic we won't be able to add this to the indexer CLI or the explorer before launch. We will for sure keep this issue open though

[koen84]

I'd like to add which network is required for indexing (mainnet, rinkeby, ropsten, kovan, etc) - as in where the data lives that the subgraph is extracting.