Phase 1 (miroir-cdo): Core Routing — Final verification complete

All Definition of Done requirements verified:
- Rendezvous determinism: ✓ (1000 randomized runs)
- Minimal reshuffling: ✓ (≤2×(1/4)×64 shards)
- Uniform distribution: ✓ (15-27 shards/node)
- Top-RF stability: ✓ (minimal change on add/remove)
- Write targets: ✓ (RG×RF nodes)
- Query distribution: ✓ (even across groups)
- Covering set: ✓ (1 node/shard + rotation)
- Merger: ✓ (all tests pass)
- Coverage: ✓ (91.80% overall, Phase 1 files 96%+)

151 tests passed in 63.04s.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

notes/miroir-cdo-verification-2026-05-09.md

@@ -0,0 +1,76 @@
+# Phase 1 (miroir-cdo): Core Routing — Final Verification Summary
+
+## Date: 2026-05-09
+
+## Definition of Done — All Requirements Met ✓
+
+### 1. Rendezvous Assignment Determinism ✓
+- **Test**: `test_rendezvous_determinism`, `acceptance_determinism_1000_runs`
+- **Result**: All 1000 randomized runs produce identical assignments for fixed (shard_id, nodes)
+
+### 2. Minimal Reshuffling on Node Add ✓
+- **Test**: `test_minimal_reshuffling_on_add`, `acceptance_reshuffle_bound_on_add`
+- **Result**: Adding 4th node moves ≤ 2 × (1/4) × 64 = 32 shards (verified)
+
+### 3. Uniform Shard Distribution ✓
+- **Test**: `test_shard_distribution_64_3_rf1`, `acceptance_uniformity_64_shards_3_nodes_rf1`
+- **Result**: 64 shards / 3 nodes / RF=1 → each node holds 15–27 shards
+
+### 4. Top-RF Placement Stability ✓
+- **Test**: `test_top_rf_stability`, `acceptance_rf2_placement_stability`
+- **Result**: Top-RF nodes change minimally on add/remove
+
+### 5. Write Targets Correctness ✓
+- **Test**: `test_write_targets_count`
+- **Result**: `write_targets` returns exactly `RG × RF` nodes, one from each group
+
+### 6. Query Group Distribution ✓
+- **Test**: `test_query_group_distribution`
+- **Result**: Queries distribute evenly across replica groups
+
+### 7. Covering Set Correctness ✓
+- **Test**: `test_covering_set_one_per_shard`, `test_covering_set_replica_rotation`
+- **Result**: One node per shard with intra-group replica rotation
+
+### 8. Merger Functionality ✓
+- **Tests**: Global sort, offset/limit, facet aggregation, score stripping, miroir field stripping
+- **Result**: All merge/facet/limit tests pass
+
+### 9. Code Coverage ≥ 90% ✓
+- **Result**: 91.80% overall line coverage for miroir-core
+  - router.rs: 96.20%
+  - topology.rs: 100.00%
+  - scatter.rs: 100.00%
+  - merger.rs: 94.67%
+
+## Test Results
+```
+test result: ok. 151 passed; 0 failed; 0 ignored; 0 measured; 0 filtered out; finished in 63.04s
+```
+
+## Implementation Summary
+
+### router.rs
+- `score(shard_id, node_id)`: Rendezvous hash with XxHash64 seed 0
+- `assign_shard_in_group`: Top-RF node selection with tie-breaking
+- `write_targets`: Returns RG × RF nodes for cross-group replication
+- `query_group`: Round-robin group selection for load balancing
+- `covering_set`: One node per shard with replica rotation
+- `shard_for_key`: Document key to shard mapping
+
+### topology.rs
+- `Topology`: Cluster nodes grouped by replica_group
+- `Node`: Health state machine (Healthy/Degraded/Joining/Draining/Failed/Active/Removed)
+- `Group`: Node container with healthy node filtering
+- Full state transition validation
+
+### scatter.rs
+- `Scatter` trait: Fan-out orchestration interface
+- `StubScatter`: Stub implementation for Phase 1
+- Request/response structures for scatter-gather
+
+### merger.rs
+- `merge`: Global sort by _rankingScore, offset/limit, facet aggregation
+- `Merger` trait: Result merge interface
+- `MergerImpl`: Production implementation
+- Stable serialization with BTreeMap