# Benchmark of marshalling performance

**URL:** <https://www.ruby-forum.com/t/benchmark-of-marshalling-performance/103402>\
**Category:** Ruby\
**Created:** [May 5, 2007, 11:35am UTC](https://www.ruby-forum.com/t/benchmark-of-marshalling-performance/103402 "2007-05-05T11:35:33Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![Sylvain\_J](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/7ab992/32.png) [@Sylvain\_J](https://www.ruby-forum.com/u/Sylvain_J)\
**Post date:** [May 5, 2007, 11:35am UTC](https://www.ruby-forum.com/t/benchmark-of-marshalling-performance/103402/1 "2007-05-05T11:35:33Z")

</div>

I had problems w.r.t. performance of Marshal.dump in one of my  
applications, so I tested if it was _that_ better to dump objects by  
sets  
instead of one by one. The results are here:

[http://www.jhaampe.org/software/marshal-perf-ruby](http://www.jhaampe.org/software/marshal-perf-ruby)

---

<div class="post-metadata">

**Author:** ![Sylvain\_J](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/7ab992/32.png) [@Sylvain\_J](https://www.ruby-forum.com/u/Sylvain_J)\
**Post date:** [May 5, 2007, 1:40pm UTC](https://www.ruby-forum.com/t/benchmark-of-marshalling-performance/103402/2 "2007-05-05T13:40:42Z")

</div>

On 5/5/07, Sylvain J. [removed\_email\_address@domain.invalid](mailto:removed_email_address@domain.invalid) wrote:

> I had problems w.r.t. performance of Marshal.dump in one of my  
> applications, so I tested if it was _that_ better to dump objects by sets  
> instead of one by one. The results are here:
> 
> ## [http://www.jhaampe.org/software/marshal-perf-ruby](http://www.jhaampe.org/software/marshal-perf-ruby)
> 
> Sylvain J.

Thanks for sharing this, it is also very comforting to know that for  
an application that does few things but _fast_ Ruby still seems fast  
enough 🙂

Salut  
Robert

---

<div class="post-metadata">

**Author:** ![Sylvain\_J](https://www.ruby-forum.com/letter_avatar_proxy/v4/letter/s/7ab992/32.png) [@Sylvain\_J](https://www.ruby-forum.com/u/Sylvain_J)\
**Post date:** [May 5, 2007, 2:05pm UTC](https://www.ruby-forum.com/t/benchmark-of-marshalling-performance/103402/3 "2007-05-05T14:05:43Z")

</div>

> Thanks for sharing this, it is also very comforting to know that for  
> an application that does few things but _fast_ Ruby still seems fast  
> enough 🙂  
> Lol. The only real problem I have with Ruby’s performance is its GC …  
> I  
> have cycles that should be around 100ms and the GC will sometime take  
> ages.