aprender-db 0.34.0

GPU-first embedded analytics database with SIMD fallback and SQL query interface
Documentation 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43

metadata:
  version: "1.0.0"
  author: "Sovereign AI Stack"
  description: "Batch query execution contract for trueno-db"
  crate: "trueno-db"
  references:
    - "Abadi et al., The Design of Modern Column-Oriented Database Systems, 2013"

equations:
  batch_throughput:
    formula: "throughput = rows_processed / elapsed_seconds"
    domain: "rows_processed ∈ ℕ, elapsed_seconds > 0"
    invariants:
      - "throughput > 0 for non-empty batches"
      - "Linear scaling: 2× rows → ~2× time"

proof_obligations:
  - type: bound
    property: "Batch query returns all matching rows"
    formal: "|result| <= |table| for any predicate"
    applies_to: all

falsification_tests:
  - id: FALSIFY-BATCH-001
    rule: "Batch correctness"
    test: "test_append_batch_olap_pattern"
    prediction: "verified"
    if_fails: "Contract violation — batch query drops rows"

kani_harnesses:
  - id: KANI-BATCH-001
    obligation: PO-BATCH-001
    property: "Batch result bounded by table size"
    bound: 8
    strategy: bounded_int
    harness: verify_batch_query_bound

qa_gate:
  id: QA-BATCH
  name: "batch query contract"
  min_coverage: 0.90
  max_complexity: 20
  required_tests: []