I have noticed recently that performance on our main app, just doesn’t render as fast as it used to in either the front end of in the editor, its slow and sluggish!
Is it just me or has anyone else seen performance drops?
1 Like
Yeah I started seeing this a couple of weeks ago. a workflow that took about 1sec to run all of the sudden started taking 8-12seconds. I just ended up slapping a loading animation on it and figured it’s the price of bubble vs going it alone.
But I’ve seen a couple of posts like this lately. I have also been somewhat reluctant to bring it up on the forums due to the “you’re doing it wrong, bubble is fast and you are slow”-crowd that turns up in all the speed related threads.
1 Like
Yep been doping this for 4 years now and if I streamline any further it will just be a static website!
I suffer from the same subject. My app, which has a professional plan + 3 capacities, takes time to do related things even when using 2-3 users. And it has really light workflows. However, most of the app is based on data usage, so I was thinking about reaching out to the support team after reviewing what I could optimize.But if my memory and experience were not mistaken, the application should fly with my current plan and additional capacity.
1 Like
Going back to this.
I’m seeing some very slow speeds now. A couple of examples.
An image has a conditional, if “input field A” contains “red” image source is “X”.
This has been instant for months. Now there’s a noticeable lag from when the text changes
to Red till the image appears. Sometimes well over 1 second. And I have made no changes to anything affecting that function for months.
Another one,
A text field with a dynamic value of “do a search for (Data type “Ball” with constraints “created by current user” and “field type: red”)Ball’s Value:Sum formatted as currency”
There are about 4 ball entries in the database it needs to look up, and this is easily taking a few seconds to do. I just don’t get it… this is with optimizing according to airdev’s speed documentation.
I think this is a problem with how the shared hosting is managed. If you have someone running a large API workflow or downloading a heap of data it seems everyone suffers. There will be some limitations on the way Bubble can manage this on a shared hosting arrangement, but dedicated costs are yet way to high at this time and hope they are working on solving this, but you are right the performance is inconsistent. I always thought it would have been better for Bubble to offer dedicated hosting through the AWS marketplace at least then we can control our own costs, location etc. However, I don’t think that is their strategy and would hope reliability, speed and scaling is the priority.
That’s fair. And I do have a fairly lengthy workflow running when setting up the user’s object, creating a ball in the example above. This was part of making stuff faster when using the app, front loading it to make it faster to pull data later on. But this workflow is not where I’m seeing the slowdowns.
I did another set up today, it does a super basic search on a handful of data and it’s taking 1-2 sec to execute. And these are the things that was instant a few months back.
So for a page where I need to see if a user has Object X and based on that show a group I now end up with a blank page for 1-2sec after the page has visually loaded everything but the group. This provides a pretty bad user experience, is confusing and generally just looks cheap.
The way I understand Bubble is, If I “do a search for” constrained to current user it will only fetch the data for the current user. And with the current user having 3 number entries to download, :sum and go “hey, I found what you were looking for”, taking 2 seconds, that does seem slow to me.
So unless you keep pulling of data to never use a constraint that isn’t Yes/No, bubble will be
very slow? But not being able to use constraints when pulling data makes the whole venture pointless, so I keep thinking, what am I missing here?
And I am pretty close to 100% sure this was near instant a few months back. I did see somewhere that as of early this year, this is now happening in the users browser and before it was being done on the bubble server, not sure if that’s true, or if that could be a possible explanation.
Had a rather interesting development on this, I thought I’d share.
That basic search I did on a handful of data that I was referring to. At the time I added
a loading bar for the 2 seconds it took to load the page. but as of yesterday it’s back
down to being near instant again, just like it was a few months back. And it’s been stable
like that for about two days now. I’m going through doing speed improvements on other
parts of the app, but the speed change was really encouraging to see. Not sure if it’s isolated
or others are seeing improved speed lately?
Yeah I have noticed this before and have submitted tickets around metrics which show maxing out one moment then ok the next in the same workflow. I still think it’s related to who else is using the resources and what processes they are running. I have a hunch at the end of the month do some metrics testing a few days out and the on the last few days of the month.
I have the same problem - on dedicated.
A very simple workflow that took a second suddenly took 8+ seconds. It’s a repeatable wf that users might do several times on 1 visit.
Filed a bug report months ago. No response/not fixed. Crickets.
Anyway, sounds like a similar issue, and as it’s appearing on dedicated I don’t think it’s capacity related.
@scriptschool as in a dedicated cluster no other user accounts on it? That sounds like a code issue. That is part of the issue you have no idea because there is no transparency in fixes or updates. Its all black box.
You still have the issue? Surprised on dedicated you don’t get priority support?