I have developed a text classification model from the AI Builder which I need to run on a large number of records (around 1m records)
I have been able to run the AI model on around 1m of records using Power Automate to loop through each record in the Dataverse table (source data) however the performance is very slow - the Power Automate flow takes around 5 days to run which is not ideal
I would like to improve the performance and reduce the execution time, I have been experimenting with various techniques such as by using parallel branches within the flow to split each branch by a period of month however it still seems to be taking a very long time!
What can I do to reduce the execution time?
Is 5 days a normal amount of time to run the AI model against around 1m records of text?
I am using a loop to go through the first 5k records in the Dataverse table, once this is complete I am using a do until loop to iterate through the remaining records (995k records) or so
Re: How to improve the performance of using Power Automate with AI Builder?
You can set the parralelism level of the "do until" by selecting "..." then settings.
That would allow to increase the number of text processed in parrallel.
Under review
Thank you for your reply! To ensure a great experience for everyone, your content is awaiting approval by our Community Managers. Please check back later.