Details
-
Improvement
-
Resolution: Fixed
-
Major
-
Cheshire-Cat
-
1
Description
sear is an in-memory bleve index implementation that is designed for efficiently executing searches against a single document
blevesearch/sear shows better performance than the adjusted in-memory version of upsidedown/moss that is used today.
Here're some unit benchmark results ..
goos: darwin
|
goarch: amd64
|
pkg: github.com/couchbase/n1fty/verify
|
cpu: Intel(R) Core(TM) i7-9750H CPU @ 2.60GHz |
test name | number of operations | ns per op | bytes per op | allocs per op |
---|---|---|---|---|
BenchmarkInMemMossIndexUpdates-12 | 30186 | 39639 | 20571 | 329 |
BenchmarkInMemMossIndexUpdateAndSearch-12 | 23836 | 49695 | 26857 | 405 |
BenchmarkInMemMossIndexUpdateSearchAndDelete-12 | 22610 | 56284 | 29889 | 429 |
BenchmarkMossWithoutOptimizations-12 | 23428 | 50219 | 26836 | 403 |
BenchmarkMossWithOptimizeReset-12 | 34096 | 33903 | 18259 | 259 |
BenchmarkMossWithOptimizeResetAndUpdate-12 | 43857 | 27402 | 17997 | 254 |
BenchmarkSearIndexUpdates-12 | 161947 | 7572 | 4800 | 93 |
BenchmarkSearIndexUpdateAndSearch-12 | 92613 | 14034 | 10116 | 152 |
BenchmarkSearIndexUpdateSearchAndDelete-12 | 72722 | 15899 | 10507 | 163 |