JamesDalton
3 years agoQrew Trainee
Pipeline speed and workarounds.
I have been testing my pipelines that I have migrated from automations and they run much slower comparatively. A process that took maybe 5 seconds max to do in my automations, now takes 40 seconds to 2 minutes.
I understand the Blaze engine was supposed to make pipelines faster, but the speed is still a far cry from the autos.
I put in a support ticket to see if this was normal for the steps the pipelines were using. The response that I got, was that it was normal and that while pipelines are slower, they scale better than autos.
Most of my pipelines are triggered by creation or update events.
A common process in my app is when a user enters in data on a parent record which gets copied to a child record. The parent record that triggered the event is updated to have the input-text cleared from the field for future use.
Here is why I did this.
Most of the users in my company are looking at many different tabs, windows, Outlook, shared folders, and remote desktop screens while they are doing their work. I wanted to make record creation and updating as convenient as possible by minimizing the amount of child records that they had to go into.
The solution that I originally had, was to have automations allow users to make all of the child records that they needed while inside of a parent record. It worked worked perfectly.
I have tried other methods like embedding child forms on the parent via iframe, but they are hard on the eyes and they do not work. There was also the option of having the child form appear as a pop-up, which I wanted to avoid as users are inundated with enough visual stimuli .
In the image below, other tabs in this record like Referrals, Notes, and Quotes, represent child records which relied on the same autos described above.
Generally, this process translated to pipelines is this:
1.) Record Update (parent record updates).
2.) Create Record (child record created).
3.) Prepare Bulk Upsert (staging parent record to have input-text cleared).
4.) Search Records (search for parent record that triggered event).
5.) For Each Record (only one record in most cases)
6.) Add Bulk Upsert Row(add blank formula-text field that clears the input)
7.) Commit Upsert (merges upsert changes to parent record)
For anyone who has migrated automations to pipelines, do you have any tips or work-arounds to make things faster?
Also, will pipeline performance improve soon?
Thank You,
James
------------------------------
James Dalton
------------------------------
I understand the Blaze engine was supposed to make pipelines faster, but the speed is still a far cry from the autos.
I put in a support ticket to see if this was normal for the steps the pipelines were using. The response that I got, was that it was normal and that while pipelines are slower, they scale better than autos.
Most of my pipelines are triggered by creation or update events.
A common process in my app is when a user enters in data on a parent record which gets copied to a child record. The parent record that triggered the event is updated to have the input-text cleared from the field for future use.
Here is why I did this.
Most of the users in my company are looking at many different tabs, windows, Outlook, shared folders, and remote desktop screens while they are doing their work. I wanted to make record creation and updating as convenient as possible by minimizing the amount of child records that they had to go into.
The solution that I originally had, was to have automations allow users to make all of the child records that they needed while inside of a parent record. It worked worked perfectly.
I have tried other methods like embedding child forms on the parent via iframe, but they are hard on the eyes and they do not work. There was also the option of having the child form appear as a pop-up, which I wanted to avoid as users are inundated with enough visual stimuli .
In the image below, other tabs in this record like Referrals, Notes, and Quotes, represent child records which relied on the same autos described above.
Generally, this process translated to pipelines is this:
1.) Record Update (parent record updates).
2.) Create Record (child record created).
3.) Prepare Bulk Upsert (staging parent record to have input-text cleared).
4.) Search Records (search for parent record that triggered event).
5.) For Each Record (only one record in most cases)
6.) Add Bulk Upsert Row(add blank formula-text field that clears the input)
7.) Commit Upsert (merges upsert changes to parent record)
For anyone who has migrated automations to pipelines, do you have any tips or work-arounds to make things faster?
Also, will pipeline performance improve soon?
Thank You,
James
------------------------------
James Dalton
------------------------------