Azure Machine Learning
An Azure machine learning service for building and deploying models.
3,333 questions
Machine Learning Workspace: Create Compute taking hours?
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(233.6, 52%, 32%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMB%3C/text%3E%3C/svg%3E)
Mark Bell
1
Reputation point
I've been having trouble with Compute for most of the day in my Azure Machine Learning workspace. I created a compute instance just fine yesterday, yet when I went to boot it this morning, it would not do so. I eventually deleted it and have tried to create a new Compute a few times since then, but it is taking FOR-EV-ER! For example, the most recent Compute started provisioning at 2:38 p.m. CST. It's now 4:38 p.m. and still not usable Compute! What's the deal?
Azure Machine Learning
{count} votes
-
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(124.80000000000001, 17%, 22%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EPP%3C/text%3E%3C/svg%3E)
Pavankumar Purilla 8,335 Reputation points Microsoft External Staff Moderator2025-06-25T05:22:11.0033333+00:00 Hi Mark Bell,
These kinds of delays are often caused by regional capacity constraints within Azure. When a compute instance is requested, Azure needs to allocate the necessary physical resources (such as specific VM sizes) in your selected region. If those resources are in high demand or temporarily unavailable especially during peak usage times or for popular VM types the provisioning process can take longer than usual, sometimes even hours. This can occur even if similar requests worked fine previously, as available capacity fluctuates throughout the day.To help resolve this, we recommend trying to create your compute instance using a different VM size, as less common sizes may have better availability. Alternatively, if your workflow allows, you might consider provisioning in a different Azure region where additional capacity may be available. If you continue to experience delays, please share the details (such as the VM size, region, and any error messages), and we’ll be happy to investigate further or escalate as needed.
-
Mark Bell 1 Reputation point
2025-06-25T13:50:37.08+00:00 Hello @Pavankumar Purilla ! I appreciate the quick response! I will try different VM size and possibly recreate the resources in another region and get back to you if I'm still having issues.. Thanks again!
-
Pavankumar Purilla 8,335 Reputation points Microsoft External Staff Moderator
2025-06-26T02:32:56.7433333+00:00 Hi Mark Bell,
Have you tried with different VM sizes and different regions? -
Pavankumar Purilla 8,335 Reputation points Microsoft External Staff Moderator
2025-06-27T03:06:24.45+00:00 Hi Mark Bell,
Have you tried using different VM sizes and regions? If you encounter any issues, please let us know. -
Mark Bell 1 Reputation point
2025-06-27T16:38:58.2566667+00:00 Hello Pavankumar. I have abandoned the usage of ML Workspace temporarily, as I began seeing a charge in our subscription for "Azure Cognitive Search" quickly rising to $255 in just a matter of a few days. I've reached out to support about this billing question. I didn't use Cognitive Search at all. Literally just cloned one Notebook from Sentinel and tried to add a Compute instance a few times.
Sign in to comment
Sign in to answer