Reassembly algorithm could be faster
Open
area: networking
area: performance
type: enhancement
- Dominant language
- Go
- Stars
- 19.3k
- Forks
- 2k
- Avg merge
- 3d 5h
- Merged PRs (30d)
- 264
Description
**Description**
The reassembler is currently O(n) speed, because we iterate through each hole linearly on each fragment we receive.
**Do you have a specific solution in mind?**
The problem is in the `updateHoles` method. Instead of using a slice to record the holes, we could use a binary search tree there.
Contributor guide
Assessment
This issue has not been assessed yet.