# Parameterized concurrency\_groups

**URL:** <https://forum.buildkite.community/t/parameterized-concurrency-groups/1743>\
**Category:** Features Requests\
**Created:** [September 29, 2021, 5:37pm UTC](https://forum.buildkite.community/t/parameterized-concurrency-groups/1743 "2021-09-29T17:37:07Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![churchmf](https://avatars.discourse-cdn.com/v4/letter/c/f05b48/32.png) [@churchmf](https://forum.buildkite.community/u/churchmf)\
**Post date:** [September 29, 2021, 5:37pm UTC](https://forum.buildkite.community/t/parameterized-concurrency-groups/1743/1 "2021-09-29T17:37:07Z")

</div>

Hello,

It would be super helpful if I could create concurrency\_groups that include environment variables or meta-data. We often have parameterized pipelines where the concurrency only matters for jobs using the same env var value. Currently all jobs in the pipeline have to wait even if a job using different (non-impacting) parameters has the lock.

 ![image](https://us1.discourse-cdn.com/flex016/uploads/buildkite1/original/1X/b251e49978cdbc5504fe144ba02c53bc62c09bb2.png)

Thanks!

---

<div class="post-metadata">

**Author:** ![paula](https://sea2.discourse-cdn.com/flex016/user_avatar/forum.buildkite.community/paula/32/954_2.png) [@paula](https://forum.buildkite.community/u/paula)\
**Post date:** [September 29, 2021, 8:23pm UTC](https://forum.buildkite.community/t/parameterized-concurrency-groups/1743/2 "2021-09-29T20:23:02Z")

</div>

Hi @churchmf!

🤔 I think you can use [dynamic pipelines](https://buildkite.com/docs/pipelines/defining-steps#dynamic-pipelines) and create a concurrency\_group for each of your situations.

Best!

---

<div class="post-metadata">

**Author:** ![churchmf](https://avatars.discourse-cdn.com/v4/letter/c/f05b48/32.png) [@churchmf](https://forum.buildkite.community/u/churchmf)\
**Post date:** [October 7, 2021, 3:24pm UTC](https://forum.buildkite.community/t/parameterized-concurrency-groups/1743/3 "2021-10-07T15:24:03Z")

</div>

Probably, but it would be nice to keep it simple and not need it. Especially since we can reference env vars in conditionals and elsewhere in a pipeline yaml. I’ll try the dynamic pipeline work around for now though, thanks!