Epic: Performance Parity with V3 #1682
Description
Objective
Close the performance gap between V3 and the current decentralized V4 network.
We’ve completed early performance exploration and now have directional signal on where V4 stands. This epic transitions from exploration to targeted improvement.
We have outlined our external commitments here https://improve.xmtp.org/t/decentralization-milestones/2039
Context
V3 represents a known performance baseline.
V4 introduces decentralization, which brings architectural benefits — but also overhead.
The goal is not to abandon decentralization tradeoffs, but to ensure that:
- Throughput is competitive with V3
- Latency is within an acceptable margin
- Stability under load matches or exceeds V3 behavior
- There are no obvious performance regressions that block adoption
Goals
- Establish a clear V3 performance baseline.
- Quantify the current V4 performance gap.
- Identify the highest-impact bottlenecks.
- Execute focused optimizations to close the gap.
- Re-test and validate improvements under comparable load.
Acceptance Criteria
- V4 reaches defined parity targets relative to V3 (throughput, latency, stability).
- Reproducible benchmarks demonstrate the improvement.
Outcome
Move from:
“V4 works, but is slower than V3.”
to:
“V4 achieves performance parity with V3 while preserving decentralization benefits.”