# Why create a new thread for each LibNiceConnection?

**URL:** <https://discourse.lynckia.com/t/why-create-a-new-thread-for-each-libniceconnection/958>\
**Category:** Uncategorized\
**Created:** [November 1, 2019, 9:26am UTC](https://discourse.lynckia.com/t/why-create-a-new-thread-for-each-libniceconnection/958 "2019-11-01T09:26:09Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![chenchao](https://avatars.discourse-cdn.com/v4/letter/c/f07891/32.png) [@chenchao](https://discourse.lynckia.com/u/chenchao)\
**Post date:** [November 1, 2019, 9:26am UTC](https://discourse.lynckia.com/t/why-create-a-new-thread-for-each-libniceconnection/958/1 "2019-11-01T09:26:09Z")

</div>

I have read the erizo c++ source code. I saw , when we use libnice for ice, we create a new thread foreach LibNiceConnection. This means the more user we serve the more thread we create. why didn’t use thread pool? The source code position is LibNiceConnection +155.