Analyzing the Performance of WebAssembly vs. Native Code

#1

Post by liminal » Tue, Nov 04, 2025, 11:13 PM UTC

Analyzing the Performance of WebAssembly vs. Native Code
ar5iv.labs.arxiv.org

Re: Analyzing the Performance of WebAssembly vs. Native Code

#2

Post by turbolent » Tue, Nov 04, 2025, 11:32 PM UTC

... in browsers. Which at best JIT compile. There are several WASM runtimes that AOT compile and have significantly better performance (e.g. ~5-10% slower).

The title is highly misleading.

Re: Analyzing the Performance of WebAssembly vs. Native Code

#3

Post by icsa » Tue, Nov 04, 2025, 11:32 PM UTC

45% slower to run everywhere from a single binary...

I'll take that deal any day!

Re: Analyzing the Performance of WebAssembly vs. Native Code

#4

Post by rlili » Tue, Nov 04, 2025, 11:33 PM UTC

That it’s not even an order of magnitude slower sounds actually pretty good!

Re: Analyzing the Performance of WebAssembly vs. Native Code

#5

Post by ModernMech » Tue, Nov 04, 2025, 11:34 PM UTC

Yeah, I've seen this when test Rust code compiled into native and wasm. I don't know about 45% though, I haven't measured it.

Re: Analyzing the Performance of WebAssembly vs. Native Code

#6

Post by vlovich123 » Tue, Nov 04, 2025, 11:36 PM UTC

This is pretty good actually considering the low hanging optimizer optimizations left and that the alternative is JS which generally performs 2-10x slower.

I think vectorization support will narrow the aggregate difference here as a lot of SPEC benefits from auto vectorization if I recall correctly.

Re: Analyzing the Performance of WebAssembly vs. Native Code

#7

Post by astafrig » Tue, Nov 04, 2025, 11:43 PM UTC

turbolent wrote: post #2

... in browsers. Which at best JIT compile. There are several WASM runtimes that AOT compile and have significantly better performance (e.g. ~5-10% slower). The title is highly misleading.

It’s not misleading to measure the performance of WebAssembly in a web browser.

Re: Analyzing the Performance of WebAssembly vs. Native Code

#8

Post by pyrolistical » Tue, Nov 04, 2025, 11:45 PM UTC

turbolent wrote: post #2

... in browsers. Which at best JIT compile. There are several WASM runtimes that AOT compile and have significantly better performance (e.g. ~5-10% slower). The title is highly misleading.

Just means the browsers can catch up.

Initially slower but then faster after full compilation

Re: Analyzing the Performance of WebAssembly vs. Native Code

#9

Post by andyferris » Tue, Nov 04, 2025, 11:46 PM UTC

icsa wrote: post #3

45% slower to run everywhere from a single binary... I'll take that deal any day!

45% slower to run everywhere from a single binary... with less security holes, without undefined behavior, and trivial to completely sandbox.

Its definitely a good deal!

Re: Analyzing the Performance of WebAssembly vs. Native Code

#10

Post by PantaloonFlames » Tue, Nov 04, 2025, 11:46 PM UTC

45% slower means..?

Suppose native code takes 2 units of time to execute.

“45% slower” is???

Would it be 45% _more time?_

What would “45% _faster_” mean?