# Issue with Parallel processing

**URL:** <https://community.wolfram.com/t/issue-with-parallel-processing/18039>\
**Category:** Questions\
**Tags:** mathematica, high-performance-computing\
**Created:** [April 29, 2021, 7:14pm UTC](https://community.wolfram.com/t/issue-with-parallel-processing/18039 "2021-04-29T19:14:37Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![twahl99](https://avatars.discourse-cdn.com/v4/letter/t/6f9a4e/32.png) [@twahl99](https://community.wolfram.com/u/twahl99)\
**Post date:** [April 29, 2021, 7:14pm UTC](https://community.wolfram.com/t/issue-with-parallel-processing/18039/1 "2021-04-29T19:14:37Z")

</div>

Hi,

I’m running Mathematica 12.0 with a license for 8 kernels. Recently when I run a parallel process (typically with Method-\>“CoarsestGrained”) only 4 of the kernels (sometimes less) are being used when it is clear that much more than half the tasks are left. In the past all 8 would be active. What is going on and how can I get all 8 kernels to be used?

Thank you,  
Troy Wahl

---

<div class="post-metadata">

**Author:** ![anon21378091](https://avatars.discourse-cdn.com/v4/letter/a/bcef8e/32.png) [@anon21378091](https://community.wolfram.com/u/anon21378091)\
**Post date:** [April 29, 2021, 9:06pm UTC](https://community.wolfram.com/t/issue-with-parallel-processing/18039/2 "2021-04-29T21:06:17Z")

</div>

Hi Troy,

Have you checked the configuration settings as described in [this](https://support.wolfram.com/27877?src=mathematica) support article? What does the following evaluate to?

```
Parallelize[Map[{#, $KernelID} &, Range[12]], Method -> "CoarsestGrained"]

```

---

<div class="post-metadata">

**Author:** ![twahl99](https://avatars.discourse-cdn.com/v4/letter/t/6f9a4e/32.png) [@twahl99](https://community.wolfram.com/u/twahl99)\
**Post date:** [April 29, 2021, 10:04pm UTC](https://community.wolfram.com/t/issue-with-parallel-processing/18039/3 "2021-04-29T22:04:39Z")

</div>

Hi,

I have checked and Parallel Kernel Configuration (local) reads:

Automatic: 8 kernels  
Number of processor cores is 8  
“Limit by license availability (8)” is checked  
“Use hard limit:16 + -” is checked

The rest in that block is unchecked.

I’ll have to run your test program when my current computation is done (could be a few days).

---

<div class="post-metadata">

**Author:** ![anon21378091](https://avatars.discourse-cdn.com/v4/letter/a/bcef8e/32.png) [@anon21378091](https://community.wolfram.com/u/anon21378091)\
**Post date:** [May 1, 2021, 7:27am UTC](https://community.wolfram.com/t/issue-with-parallel-processing/18039/4 "2021-05-01T07:27:27Z")

</div>

Hi,

I just ran your code and got:

```
In[30]:= Parallelize[Map[{#, $KernelID} &, Range[12]], 
 Method -> "CoarsestGrained"]

Out[30]= {{1, 1}, {2, 1}, {3, 2}, {4, 2}, {5, 3}, {6, 3}, {7, 4}, {8, 
  4}, {9, 5}, {10, 6}, {11, 7}, {12, 8}}

```

Which is as it should be, but the things I’m running are huge (functional arrays 100’s wide, and over 30,000 long) yet not all the kernels are involved…