andresakata (André Guimarães Sakata)
- Login: andresakata
- Email: andregsakata@gmail.com
- Registered on: 07/29/2018
- Last sign in: 03/09/2024
Issues
| open | closed | Total | |
|---|---|---|---|
| Assigned issues | 0 | 0 | 0 |
| Reported issues | 0 | 0 | 0 |
Activity
07/30/2018
-
10:46 PM Ruby Feature #14718: Use jemalloc by default?
- davidtgoldblatt (David Goldblatt) wrote:
> I don't think this benchmark is a useful way to compare performance between versions 3 and 5 of jemalloc. In between them was the advent of time-based purging, where the allocator waits for a w... -
- Got it, thanks @fweimer.
I did run a test with **jemalloc 5.1** and I've got the same results as you did.
```
Min. :242.9
1st Qu.:1593.5
Median :1737.7
Mean :1691.5
3rd Qu.:1860.0
Max. :2127.9
```
Remembering my pre... -
- Hello @fewimer, very interesting.
Just to understand one thing, were you using MALLOC_ARENA_MAX=2 in your first test?
07/29/2018
-
- bluz71 (Dennis B) wrote:
> @andresakata,
> ...
I had forgotten to answer it. Yes, in that test I was using jemalloc 3.6.0. -
- I removed the NewRelic dependency and it's using the `get_process_mem` gem instead. It should work on different platforms as well.
Now there is a README file in the repository that may help you to reproduce the tests. Let me know you ... -
- fweimer (Florian Weimer) wrote:
> Thanks. Would it be possible to eliminate the dependency on `newrelic_rpm`, and possibly Rails as well? We don't have a commercial `newrelic_rpm` license, and we are not Ruby developers, so this is no... -
- Hi!
I'm another Ruby user that used to have memory bloat problems and switched to `jemalloc` as well.
I just wrote a simple script (36 lines) that seems to reproduce the issue.
https://github.com/andresakata/ruby-memory-bloat
...