mat/simdnbt
1
0
Fork 0
mirror of https://github.com/azalea-rs/simdnbt.git synced 2025-08-02 07:26:04 +00:00
Code Activity Actions

update table in readme

Browse source
This commit is contained in:
mat 2024-05-12 23:57:02 +00:00
parent 5ffce1a309
commit a6f47171bf
2 changed files with 10 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
simdnbt/README.md
View file

@ -60,10 +60,12 @@ The most significant and simple optimization you can do is switching to an alloc
## Implementation details
Simdnbt currently makes use of SIMD instructions for two things:
- swapping the endianness of int arrays
- checking if a string is plain ascii for faster mutf8 to utf8 conversion
- swapping the endianness of int arrays
- checking if a string is plain ascii for faster mutf8 to utf8 conversion
Simdnbt ~~cheats~~ takes some shortcuts to be this fast:
1. it requires a reference to the original data (to avoid cloning)
2. it doesn't validate/decode the mutf-8 strings at decode-time
@ -75,7 +77,7 @@ Here's a benchmark comparing Simdnbt against a few of the other fastest NBT crat
| Library | Throughput |
| --------------------------------------------------------------------------- | ------------ |
| [simdnbt::borrow](https://docs.rs/simdnbt/latest/simdnbt/borrow/index.html) | 717.45 MiB/s |
| [simdnbt::borrow](https://docs.rs/simdnbt/latest/simdnbt/borrow/index.html) | 1.0900 GiB/s |
| [simdnbt::owned](https://docs.rs/simdnbt/latest/simdnbt/owned/index.html) | 329.10 MiB/s |
| [shen_nbt5](https://docs.rs/shen-nbt5/latest/shen_nbt5/) | 306.58 MiB/s |
| [azalea_nbt](https://docs.rs/azalea-nbt/latest/azalea_nbt/) | 297.28 MiB/s |
@ -85,14 +87,13 @@ Here's a benchmark comparing Simdnbt against a few of the other fastest NBT crat
| [hematite_nbt](https://docs.rs/hematite-nbt/latest/nbt/) | 108.91 MiB/s |
And for writing `complex_player.dat`:
| Library | Throughput |
| Library | Throughput |
| ----------------| ------------ |
| simdnbt::borrow | 2.5914 GiB/s |
| azalea_nbt | 2.1096 GiB/s |
| simdnbt::owned | 1.9508 GiB/s |
| azalea_nbt | 2.1096 GiB/s |
| simdnbt::owned | 1.9508 GiB/s |
| graphite_binary | 1.7745 GiB/s |
The tables above were made from the [compare benchmark](https://github.com/azalea-rs/simdnbt/tree/master/simdnbt/benches) in this repo.
Note that the benchmark is somewhat unfair, since `simdnbt::borrow` doesn't fully decode some things like strings and integer arrays until they're used.
Also keep in mind that if you run your own benchmark you'll get different numbers, but the speeds should be about the same relative to each other.

4
simdnbt/benches/compare.rs
View file

@ -138,9 +138,9 @@ fn bench(c: &mut Criterion) {
// bench_read_file("hello_world.nbt", c);
// bench_read_file("bigtest.nbt", c);
// bench_read_file("simple_player.dat", c);
// bench_read_file("complex_player.dat", c);
bench_read_file("complex_player.dat", c);
// bench_read_file("level.dat", c);
bench_read_file("inttest1023.nbt", c);
// bench_read_file("inttest1023.nbt", c);
}
criterion_group!(compare, bench);