This repository has been archived by the owner on Nov 6, 2020. It is now read-only.
Blooms recomputation in case of a fork is very slow #8552
Labels
F7-optimisation 💊
An enhancement to provide better overall performance in terms of time-to-completion for a task.
M4-core ⛓
Core client code / Rust.
Milestone
Node list
I am sending externally created transactions to private PoA chain with eth_sendRawTransaction.
Send 6400 transactions from one account at a time and wait for synchronization.
Each transaction uses 1649 bytes and uses approximately 1.15 M gas.
If a temporary fork occurs during this sync, subsequent synchronization will be extremely slow.
In normal case, it takes about 7 minutes for all transactions to be processed.
If a temporary fork occurs, it takes about 30 minutes for all transactions to be processed.
During this time, the CPU usage rate will be 100%, but memory and storage are calm.
In this log, a temporary fork occurs between AU1 and AU2 at #1638747, and long time synchronization of OB1 starts at that timing.
At this time, there were about 3700 pending transactions for OB1.
AU1
AU2
OB1
The text was updated successfully, but these errors were encountered: