swath run trace
Run
The guesses, and how they turned out
Before any range lists, swath asks the store where the folders are — a few
delimiter=/ listings of the top of the tree — and cuts the keyspace at them into
starting ranges. A folder name says nothing about how many objects are behind it, so this is
a guess, and it is the only cheap information the engine will get.
Seed ranges, by objects actually found behind them
Objects by top-level directory
Grouped by first path segment, with runs of six or more digits generalized so dated sibling directories collapse into one row — a grouping choice for this table only.
What the run actually felt like
Objects listed per second, and live ranges, over the run
Most of these keys do not exist
To split a range swath needs a boundary key, and it does not need one that exists — a listing can start after any key. The pivots below are verbatim from this run; most are synthesized coordinates, not object names.
| Mechanism | Splits | Sample pivots, verbatim |
|---|
The whole run, on one map
Horizontal is the keyspace weighted by objects — equal width means an equal number of objects — and vertical is time running downward. Each rectangle is one range over the slice it owned and the time it owned it, colored by which original seed guess it descends from. Because width is objects, a rectangle's area is roughly the work it did.
Wide and short is a big slice eaten fast. A tall thin column is a serial tail — a dense region that could not be divided, outliving everything around it.
Watch it happen
The same events replayed in order, on the same axis. Bands are live ranges, filled to each worker's cursor; ticks are split pivots as they commit. A split ends its range at the cursor and hands the rest on as two new ranges.
The ledger
Run record
Provenance
Fields absent from the retained run-facts record read "unknown in retained
evidence" rather than being inferred. The machine-readable copy is in
#swath-run-facts.