gianm opened a new pull request #127: URL: https://github.com/apache/datasketches-memory/pull/127
It is faster by 3–10x, at least on my machine (an Intel MacBook Pro). Some benchmarks comparing `clear()` vs `fill(0)` after this patch: ``` Benchmark (numBytes) (where) Mode Cnt Score Error Units MemoryBenchmark.clear 4 offheap avgt 15 10.706 ± 0.166 ns/op MemoryBenchmark.clear 16 offheap avgt 15 11.031 ± 0.552 ns/op MemoryBenchmark.clear 64 offheap avgt 15 10.572 ± 0.162 ns/op MemoryBenchmark.clear 256 offheap avgt 15 11.539 ± 0.081 ns/op MemoryBenchmark.clear 1024 offheap avgt 15 17.520 ± 0.067 ns/op MemoryBenchmark.clear 1025 offheap avgt 15 30.936 ± 0.537 ns/op MemoryBenchmark.clear 4096 offheap avgt 15 73.675 ± 1.968 ns/op MemoryBenchmark.clear 16384 offheap avgt 15 298.331 ± 5.742 ns/op MemoryBenchmark.clear 65536 offheap avgt 15 1456.163 ± 51.568 ns/op MemoryBenchmark.clear 262144 offheap avgt 15 7050.704 ± 268.609 ns/op MemoryBenchmark.fill 4 offheap avgt 15 117.083 ± 0.723 ns/op MemoryBenchmark.fill 16 offheap avgt 15 117.528 ± 1.009 ns/op MemoryBenchmark.fill 64 offheap avgt 15 121.382 ± 2.037 ns/op MemoryBenchmark.fill 256 offheap avgt 15 131.476 ± 1.351 ns/op MemoryBenchmark.fill 1024 offheap avgt 15 182.377 ± 1.232 ns/op MemoryBenchmark.fill 1025 offheap avgt 15 145.153 ± 2.790 ns/op MemoryBenchmark.fill 4096 offheap avgt 15 363.053 ± 2.628 ns/op MemoryBenchmark.fill 16384 offheap avgt 15 1205.843 ± 47.600 ns/op MemoryBenchmark.fill 65536 offheap avgt 15 4513.897 ± 182.637 ns/op MemoryBenchmark.fill 262144 offheap avgt 15 17750.584 ± 1029.368 ns/op ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
