You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: README.md
+10-6Lines changed: 10 additions & 6 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -380,19 +380,23 @@ This allows custom serialization while maintaining a generic interface for the l
380
380
381
381
## Performance
382
382
383
-
The `json-as` library has been optimized to achieve multi-GB/s processing speeds through SIMD and SWAR acceleration and highly efficient transformations. Below are detailed statistics comparing performance metrics such as build time, operations-per-second, and throughput.
383
+
The `json-as` library is engineered for **multi-GB/s processing speeds**, leveraging SIMD and SWAR optimizations along with highly efficient transformations. The charts below highlight key performance metrics such as build time, operations-per-second, and throughput.
384
+
385
+
You can **re-run the benchmarks** at any time by clicking the button below. After the workflow completes, refresh this page to view updated results.
This chart compares JSON-AS (both SWAR and SIMD variants) to JavaScript's built-in JSON implementation. Benchmarks were run in a GitHub Actions environment--on most modern machines, you may see even higher throughput.
391
+
The following charts compare JSON-AS (both SWAR and SIMD variants) against JavaScript's native `JSON` implementation. Benchmarks were conducted in a GitHub Actions environment. On modern hardware, you may see even higher throughput.
388
392
389
-
You can re-run the benchmarks anytime by clicking the button below, waiting for it to complete, and then refreshing your page here.
393
+
> Note: Benchmarks reflect the **latest version**. Older versions may show different performance.
> Note: I have focused on extensively optimizing serialization. I used to have deserialization be highly unsafe and extremely fast, but I've since doubled down on safety for deserialization which has negatively affected performance. I will be optimizing soon.
0 commit comments