Hacker News Reader

Comment by sod | Hacker News Reader

While annoying, all intellij EAPs (early access previews) are very cpu intensive, as they run in debug mode by default to create diagnostics and crash reports. Kinda similar like if you would keep the debugger open in your browser. Not sure if you can disable it, but thats the main reason why I stopped using EAPs for work.

But the ram is typical intellij, they take what they can get from the OS :)

View on HN

> But the ram is typical intellij, they take what they can get from the OS :)

Aha, thanks! I was wondering about that (using CLion 2021.3 with Rust plugin with currently a tiny test project open, currently using 2.3 GiB RAM, puah)... .

Well, they cache many things to make code analysis fast, so they are one of the few softwares that do make good use of memory. Some older version had very low heap size set and that can cause slow downs, so 2+ GB is recommended at least in my experience.

i give intellij about 10GB to play with. i don't care. im not using it fot anything else. cache away. will happily trade ram for speed

> they take what they can get from the OS :

It's the opposite, they run in JVM so there's an upper bound at around -Xmx ("around" because you still have stack allocated and JNI allocated, but heap is usually the biggest pool of memory by a large margin)

on the flip side it also has a lower bound at -Xms

Personally I have mine running with 16/32G heap. It's my work computer not like I need the 64G ram for gaming :)

any source on that claim? I.e. that eap are cpu intensive?

I can't find a source to my claim. Just now downloaded webstorm 2022.2.3 stable and webstorm 2022.3 EAP to compare, but they seem to behave similar in cpu. Every now and then in the EAP in the activity monitor a "diagnostics dumper" pops up, but only for a brief moment. Maybe they improved it a lot. I remember complaining about EAP performance a long time ago and remembered an answer to expect a few more cpu usage from an EAP (the diff was not as stark as 100% though, it was more like 20% EAP vs 5% stable).

I mean please don't judge the CPU profile of an EAP to the final product.

I don't have a source other than that I develop plugins for IntelliJ professionally, but it's definitely true that performance is generally not a focus earlier in EAP cycles. I think they mention it when you download an EAP release.

Reply on news.ycombinator.com