Optimize bandwidth aggregation with out-of-order processing and dynamic allocation

[devin-ai-integration]

Bandwidth Aggregation Optimization

This PR implements two key optimizations to improve bandwidth aggregation efficiency when using workers with unbalanced bandwidth configurations:

Changes Made

1. Out-of-order Frame Processing: Modified the receiver to buffer frames that arrive out of order, preventing slower workers from blocking faster ones
2. Dynamic Frame Allocation: Implemented performance tracking and dynamic frame distribution based on actual worker throughput

Performance Results

• Before: 100KB/s + 500KB/s workers achieved only 34.72% efficiency (208.31KB/s actual vs 600KB/s expected)
• After: 100KB/s + 500KB/s workers achieved 69.44% efficiency (416.64KB/s actual vs 600KB/s expected)
• Balanced Workers: 300KB/s + 300KB/s workers achieved 100.59% efficiency (603.53KB/s actual vs 600KB/s expected)

Testing

Tested with bandwidth-test tool using various worker configurations to verify improvements:

• Unbalanced workers (100KB/s + 500KB/s): Efficiency improved from 34.72% to 69.44%
• Balanced workers (300KB/s + 300KB/s): Achieved 100.59% efficiency

Link to Devin run: https://app.devin.ai/sessions/3571e39c6d55434a84d80937d3437d5b
Requested by: fatedier

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment and CI monitoring