Synapse Spark Pool starting time is significant longer than before

Question

Synapse Spark Pool starting time is significant longer than before

Z.K.Z 221

Hello,

Over the past few months, I’ve noticed that Synapse Spark Pools take significantly longer to start. Previously, it took around 5 minutes to start a Spark Pool from scratch, but now it takes about 10 minutes.

Is there any way to reduce the startup time?

Smaran Thoomu 32,155 Reputation points Microsoft External Staff Moderator

2025-10-24T07:25:17.99+00:00

Z.K.Z We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.

1 answer

Your answer

Smaran Thoomu 32,155 Reputation points Microsoft External Staff Moderator

2025-10-24T07:25:17.99+00:00

Z.K.Z We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.

Answer 1

Hi Z.K.Z
Thank you for reaching out.

You’re right -recently, many users have observed that Synapse Spark Pools take longer to start, sometimes increasing from about 5 minutes to around 10 minutes or more. This can happen due to recent Spark version updates, higher regional demand, or backend resource provisioning changes. Microsoft is aware of this behavior and continues to work on optimizing the startup experience.

Here are a few things you can try to help reduce startup delays:

Try Spark version 3.4 instead of 3.5 - some users have reported slightly faster pool initialization times on earlier versions.
Reduce the size or number of attached custom libraries/packages - large or complex dependencies can slow down startup.
Keep your pool “warm” - increase the idle timeout setting so the pool doesn’t fully shut down between jobs. This helps if you run workloads frequently.
Monitor Azure updates - performance improvements are being rolled out gradually. If startup time continues to worsen, consider raising a support ticket for further investigation.

Currently, there’s no direct configuration available to fully control Spark Pool cold-start time, but the above steps should help minimize the delay.

Hope this helps!

Share via

Synapse Spark Pool starting time is significant longer than before

1 answer

Your answer