Repository navigation
Conversation
Signed-off-by: Daniel King <dan@spiraldb.com>
… timing Signed-off-by: Daniel King <dan@spiraldb.com>
Signed-off-by: Daniel King <dan@spiraldb.com>
Signed-off-by: Daniel King <dan@spiraldb.com>
Signed-off-by: Daniel King <dan@spiraldb.com>
Merging this PR will improve performance by 11.23%1 warning
Performance Changes
Tip Curious why performance improved? Comment Comparing |
Signed-off-by: Daniel King <dan@spiraldb.com>
Signed-off-by: Daniel King <dan@spiraldb.com>
|
@claude review |
|
Claude finished @danking's task in 59s —— View job Review: PCO selection benchmarksThis adds benchmarks only, with no production code. The oracle checks before each timed loop are good: every workload is checked for correctness against the uncompressed Should fix before un-drafting
Smaller suggestions
Unrelated CodSpeed regressionsThe six regressions in the CodSpeed comment are in ChecksI only read the code. I did not run clippy, fmt or the benches; your CLAUDE.md leaves checks to you, and the PR already notes that scoped compile and lint are pending. |
Signed-off-by: Daniel King <dan@spiraldb.com>
…load Signed-off-by: Daniel King <dan@spiraldb.com>
Signed-off-by: Daniel King <dan@spiraldb.com>
|
@claude review |
|
Claude finished @danking's task in 51s —— View job Review: PCO selection benchmarks
Nothing has changed since the last review. HEAD is still Corrections to the earlier review
New, smaller points
ChecksI only read the code. I did not run clippy, fmt or the benches; CLAUDE.md leaves verification to you, and the PR notes that scoped compile, lint and cloud calibration are still pending. |
Signed-off-by: Daniel King <dan@spiraldb.com>
|
Addressed the benchmark review in e599953: cache each primitive/PCO fixture once with LazyLock and force setup before the runner, use the existing MiMalloc dependency, label nullable/non-nullable and defined/missing cases, name the workload constants, reject unknown selection names, and remove the unnecessary scalar drop. The retained case is explicitly documented as single-page; scalar.rs supplies the scattered control. The fixed workload sizes are unchanged. CPU Simulation numbers are not wall-time calibration, and the allocator/setup changes have not been rebenchmarked. The repository-wide optimized-out warning was not attributed to these cases, so no benchmark was removed or disabled. Local independent source review passed; no builds, tests, formatting commands, or benchmarks were run. |
|
This PR has been marked as stale because it has been open for 14 days with no activity. Please comment or remove the stale label if you wish to keep it active, otherwise it will be closed in 7 days |
Add public PCO benchmarks for decoding, metadata opening, take, slice, range selection, and scalar probes, using fixed workloads and untimed correctness checks. These establish a comparable baseline for the page-index optimizations using the existing benchmark and CodSpeed infrastructure.