SPB Git

Repository Public

BENCHMARKS.md 4.5 KB

# SheetData benchmarks

Measured with orix-data-bench (this package). These are measurements of one run on one machine, not guarantees; every number moves with hardware, disk cache state, thread count and data shape.

# Setup

Date 2026-11-21
Machine Mac17,6, Apple M5 Max, 18 cores, 48 GiB RAM, macOS 27.0 (26A5425a)
Engine DuckDB v2.0.1 (vendored universal dylib), 18 threads, memory_limit 24 GiB (default = half of RAM), release build
Dataset 10,000,000 synthetic rows × 8 columns: id BIGINT, category VARCHAR (20 values), region VARCHAR (8), order_date DATE (2020–2024), amount DOUBLE, quantity BIGINT (1–50), customer_id BIGINT (100,000 values), status VARCHAR (5)
Files CSV 559.9 MB (written by the tool, streamed through a 4 MiB buffer), Parquet 108.1 MB (zstd, written by DuckDB from the CSV), plus a 100,000-row customers CSV for the join
Cache state the files were written seconds before being read, so they were in the OS page cache: import and scan times are CPU-bound, not disk-bound

Command:

sh
swift build -c release
.build/release/orix-data-bench --dir /tmp/orix-bench --rows 10000000 > report.json

Queries are run three times; the table gives the median and the best run. One-off steps (generation, import, profile, writes to the grid) are run once.

# Results

Step Median (s) Best (s) Notes
Generate CSV (Swift, streamed) 3.931 10,000,000 rows, 559.9 MB, process RSS 15 MiB
Register CSV as a lazy view 0.230 sniffing only, no rows loaded
Aggregate straight from the CSV view (GROUP BY region) 0.138 0.129 full parse of the file each time
Convert CSV → Parquet (exportQuery) 0.671 108.1 MB
Import CSV into a table (importFile) 0.300 10,000,000 rows, ≈ 33 M rows/s
Row count 0.0004 0.0003
Filtered aggregation (2 IN values, date ≥, amount >, GROUP BY category) 0.014 0.012 20 groups
Pivot region × category by year, sum + % of grand total (PivotEngine.compute) 0.038 0.037 169 row items × 6 column items; DuckDB returned 1,014 grouped rows
Join with customers on customer_id + GROUP BY segment, status (transform pipeline) 0.088 0.065 20 groups
Preview window offset 5,000,000, limit 200 (insertion order) 0.005 0.003
Preview window sorted by amount DESC, offset 1,000, limit 200 0.033 0.032 top-N with a deterministic tie-break
Preview window with a filter, offset 10,000, limit 200 0.007 0.006
Profile of the 8 columns (+ top 3 values each) 1.127 exact quantiles on 10 M rows dominate
Pivot computed, rendered and written to a sheet 0.146 171 × 14 cells, one undo step
Materialize a 100,000-row × 8-column extract into a sheet 0.925 800,008 cells through CalcEngine.set, one undo step
Materialize the whole table into a sheet refused exceedsSheetLimits: 10,000,001 rows × 8 columns do not fit 1,048,576 × 16,384
Parquet, pushdown filter (id BETWEEN 9,000,000 AND 9,010,000) 0.004 0.004 row groups skipped from statistics; 10,001 rows matched
Parquet, projection + aggregate (GROUP BY category, 2 of 8 columns read) 0.011 0.010
Parquet, the filtered aggregation above 0.014 0.014 no import needed

Peak resident memory of the whole process (getrusage, ru_maxrss): 2,941 MiB, including the imported table (in-memory database), the 800,008-cell worksheet and the 1,048,576 rows fetched before the full-table materialization was refused. After the import alone it was 1,628 MiB.

Consistency check at the end of the run: the table, the Parquet view and the CSV view all hold 10,000,000 rows and the same total amount (compared as integer cents).

Two earlier runs of the same binary on the same machine gave 0.297 s and 0.364 s for the import, 0.011 s for the filtered aggregation, 0.029–0.031 s for the pivot, 0.028–0.029 s for the join and 0.003–0.004 s for the middle preview window: expect that kind of spread.

# Reading the numbers

  • The grid never sees the 10 M rows: the UI reads 200-row windows (preview), and only bounded results (a pivot, an extract) cross into a worksheet.
  • Writing to the grid is the slow door: about 0.86 M cells/s through SheetCore's per-cell set, against tens of millions of rows per second inside the engine.
  • swift test does not run this benchmark. A 1,000,000-row end-to-end test exists behind ORIX_DATA_LARGE_TESTS=1.
Files main
Menu

Theme