Implement the RPC v2 performance benchmark suite (data profiles + test scenarios)

[tamirms]

TL;DR

Implement the benchmark suite specified in the RPC v2 Performance Evaluation doc — §6 data profiles and §7 test scenarios — to verify the §4 SLAs and catch regressions.

Scope

• §6 data profiles — the deterministic synthetic generator and its three profiles (pubnet-current, pubnet-projected-2×, event-stress).
• §7 test scenarios — the query (frozen + hot) and ingestion (frozen + hot) scenarios with their knob sweeps, plus the §7.0 methodology (cold-cache eviction, hot-tier warmup, round counts, percentile stats).
• Re-evaluate the scenarios against what already exists and reuse what fits from bench(fullhistory): synthetic-ledger datasets via stellar-core apply-load as an ingest/query bench source #762 (synthetic data) and Full-history benchmark commands — ingest + query across all data types #771 (benchmark commands) rather than rebuilding.

Reference

RPC v2 — SLAs and Test Scenarios — §6, §7 (targets in §4).

Related

• Run full-history benchmark commands in CI to catch perf regressions #775 — runs these in CI for regression detection.