Memory spikes in Analysis Services

Question

Memory spikes in Analysis Services

Billy Duong 5

We've recently been experiencing large memory spikes while processing a few of our tabular models on different Azure Analysis Services instances.

There were no major changes to the data, so we were not able to determine why it was happening.

The issue then went away on 01/26 after a week of doing so. This coincided with a maintenance window on Analysis Services in the West US region (the region we use).

Anyone else experience similar issue and/or know if the maintenance was what actually resolved this or just coincidental timing?

phemanth 15,320 Reputation points Microsoft External Staff

2025-01-30T04:41:23.1333333+00:00
@Billy Duong

Thanks for the question and using MS Q&A platform.

It Seems like you've been dealing with some frustrating memory spikes in Azure Analysis Services. It's good to hear that the issue resolved itself after the maintenance window, but I understand why you'd want to know if the maintenance was the actual fix or just a coincidence.

Memory spikes in Azure Analysis Services can be caused by various factors, including complex DAX calculations, large data volumes, or inefficient data models

Maintenance windows can sometimes resolve these issues if they involve updates or optimizations to the underlying infrastructure.

Here are steps you can take to monitor and optimize memory usage in Azure Analysis Services:

Monitor Memory Usage: Use the Azure portal to monitor memory usage and identify any patterns or specific times when spikes occur.

Optimize Data Models: Simplify your data models and reduce the number of calculated columns and tables if possible.

Review DAX Queries: Ensure that your DAX queries are optimized and not causing excessive memory consumption.

Scale Up Resources: If memory issues persist, consider scaling up your Azure Analysis Services instance to a higher tier with more memory

I hope the above steps will resolve the issue, please do let us know if issue persists. Thank you
phemanth 15,320 Reputation points Microsoft External Staff

2025-01-31T01:29:57.5466667+00:00

@Billy Duong 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.
Billy Duong 5 Reputation points

2025-01-31T18:42:58.5333333+00:00

Thanks for the response. No resolution has been found. We've previously done all the steps you suggested while it was occurring and it didn't help - other than scaling up resources, of course. But that wasn't really necessary once the spike went away.

One of the affected models was fairly simple - not many objects/tables/calculations in it. Not really any DAX queries in it. But the spike was still occurring on it.
Nahuel Beni 0 Reputation points

2025-02-01T19:26:47.9433333+00:00

Hi Billy, we had a similar issue with a system last week. Eventually it was solved, but we are now seeing it again.

Last week we "workaround" it by moving to another Azure region but today that's not the case. Are you back to facing the issue?
phemanth 15,320 Reputation points Microsoft External Staff

2025-02-03T12:28:58.9566667+00:00
@Billy Duong

Thanks for information

Given that you've already tried the common optimization steps, it please consider few additional approaches:

Check for Background Processes: Sometimes, background processes or other services running on the same infrastructure can cause unexpected memory spikes. Ensure there are no other processes that might be interfering.

Review Maintenance Logs: Look into the maintenance logs around the time the issue resolved itself. There might be clues indicating what changes were made during the maintenance window.

Analyze Query Patterns: Even if there aren't many DAX queries, it's worth analyzing any query patterns or user interactions that might coincide with the memory spikes.

I hope the above steps will resolve the issue, please do let us know if issue persists. Thank you

Your answer

phemanth 15,320 Reputation points Microsoft External Staff

2025-01-30T04:41:23.1333333+00:00

@Billy Duong

Thanks for the question and using MS Q&A platform.

It Seems like you've been dealing with some frustrating memory spikes in Azure Analysis Services. It's good to hear that the issue resolved itself after the maintenance window, but I understand why you'd want to know if the maintenance was the actual fix or just a coincidence.

Memory spikes in Azure Analysis Services can be caused by various factors, including complex DAX calculations, large data volumes, or inefficient data models

Maintenance windows can sometimes resolve these issues if they involve updates or optimizations to the underlying infrastructure.

Here are steps you can take to monitor and optimize memory usage in Azure Analysis Services:

Monitor Memory Usage: Use the Azure portal to monitor memory usage and identify any patterns or specific times when spikes occur.

Optimize Data Models: Simplify your data models and reduce the number of calculated columns and tables if possible.

Review DAX Queries: Ensure that your DAX queries are optimized and not causing excessive memory consumption.

Scale Up Resources: If memory issues persist, consider scaling up your Azure Analysis Services instance to a higher tier with more memory

I hope the above steps will resolve the issue, please do let us know if issue persists. Thank you
phemanth 15,320 Reputation points Microsoft External Staff

2025-01-31T01:29:57.5466667+00:00

@Billy Duong 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.
Billy Duong 5 Reputation points

2025-01-31T18:42:58.5333333+00:00

Thanks for the response. No resolution has been found. We've previously done all the steps you suggested while it was occurring and it didn't help - other than scaling up resources, of course. But that wasn't really necessary once the spike went away.

One of the affected models was fairly simple - not many objects/tables/calculations in it. Not really any DAX queries in it. But the spike was still occurring on it.
Nahuel Beni 0 Reputation points

2025-02-01T19:26:47.9433333+00:00

Hi Billy, we had a similar issue with a system last week. Eventually it was solved, but we are now seeing it again.

Last week we "workaround" it by moving to another Azure region but today that's not the case. Are you back to facing the issue?
phemanth 15,320 Reputation points Microsoft External Staff

2025-02-03T12:28:58.9566667+00:00

@Billy Duong

Thanks for information

Given that you've already tried the common optimization steps, it please consider few additional approaches:

Check for Background Processes: Sometimes, background processes or other services running on the same infrastructure can cause unexpected memory spikes. Ensure there are no other processes that might be interfering.

Review Maintenance Logs: Look into the maintenance logs around the time the issue resolved itself. There might be clues indicating what changes were made during the maintenance window.

Analyze Query Patterns: Even if there aren't many DAX queries, it's worth analyzing any query patterns or user interactions that might coincide with the memory spikes.

I hope the above steps will resolve the issue, please do let us know if issue persists. Thank you

Share via

Memory spikes in Analysis Services

Your answer