tapdb: allow batch inserting proofs when syncing universe, remove write lock for RegisterIssuance #449
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
guggero
force-pushed
the
remove-mint-lock
branch
2 times, most recently
from
91848b3 to
a7d03b6
Compare
Roasbeef
requested changes
Aug 15, 2023
guggero
force-pushed
the
remove-mint-lock
branch
from
a7d03b6 to
e9d156c
Compare
|
I was able to complete a full asset mint with 10k assets with Postgres on my local machine today 💯 Here are some statistics of the run: EDIT: After a few optimizations, I was able to get a full run in 4541s. Attaching the log if anyone is curious: |
guggero
force-pushed
the
remove-mint-lock
branch
2 times, most recently
from
e308cdc to
711414b
Compare
|
Still want to add some additional unit tests to the database around batched insertion, but the PR should otherwise be ready for re-review. |
guggero
force-pushed
the
remove-mint-lock
branch
from
eab0b65 to
66effd0
Compare
Roasbeef
reviewed
Aug 23, 2023
jharveyb
reviewed
Aug 29, 2023
|
Needs a rebase! |
guggero
force-pushed
the
remove-mint-lock
branch
from
66effd0 to
234008e
Compare
We avoid several encode/decode cycles by keeping the raw proof in the minting leaf instead of the encoded one.
When syncing proofs from a remote universe, we download the proofs in parallel. But to avoid also hammering the DB with parallel write requests, we batch the writes.
guggero
force-pushed
the
remove-mint-lock
branch
from
234008e to
df11d35
Compare
We also want to insert proofs in batches into the local universe when we mint multiple assets.
Instead of having three make goals for the optional itests, we use an argument (make itest optional=true), which also makes it easier for us to increase test and postgres harness timeouts during those optional tests.
It seems like Ubuntu 22.04 (current latest) terminates build jobs because of high CPU usage sometimes (see actions/runner-images#6680). We attempt to fix this by using the older runner.
guggero
force-pushed
the
remove-mint-lock
branch
from
df11d35 to
beb76bc
Compare
Roasbeef
approved these changes
Sep 7, 2023
|
|
||
| // We use an error group to simply the error handling of a goroutine. | ||
| batchSyncEG.Go(func() error { | ||
| newLeafProofs, err = s.batchStreamNewItems( |
There was a problem hiding this comment.
Very cool pattern! I think we might need just one more light layer of abstraction to make it a bit more readable, but I dig the structured concurrency here.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #374.
Opening as draft, want to address the following TODOs before full review:
Code cleanupAllow federation envoy to also support batch inserting, will speed up local insertion