search

hirosystems / ordhook

Build indexers, standards and protocols on top of Ordinals and Inscriptions (BRC20, etc).
Apache License 2.0
190 stars 56 forks source link

Speed-up ordinals indexing #124

Closed mwalol closed 1 year ago

mwalol commented 1 year ago

having the bitcoin node and the chainhook on the same machine we should have a flag dev to read the blockchain data from the disk instead of rpc for fast syncing once's synced we have choice to read from rpc

lgalabru commented 1 year ago

Hey @mwalol! Thanks for opening this issue, this is indeed a concern on my end, and we'll be doing a new pass on indexing speed in the next weeks.

lgalabru commented 1 year ago

Next iteration: Pass 1: Download and compress blocks (out of order) Pass 2: Identify Inscriptions and Compute ordinal numbers (out of order) Pass 3: Compute Transfers, Inscription numbers, identify eventual curse types (in order)

smcclellan commented 1 year ago

In scope for this ticket is Pass 1 from above.

sabbyanandan commented 1 year ago

Hi, @mwalol! Your speed-relatead observation and the specifics behind how you're trying to use hord made be curious. I would love to learn more about your use of hord and use cases.

Can you point me to any details where I can learn/educate myself? It'll also help us think through other feature-toggles and the overall hord's DevEx. Alternatively, I would love to also talk and exchange ideas: https://calendly.com/sabbyanandan/platform-office-hours

lgalabru commented 1 year ago

A bunch of changes recently merged in develop. Speed will always be an area of R&D, but the next lowest hanging fruit might be on devops side.