Forum Discussion
MichaelTamoush
Qrew Captain
Evan,
This seems like a boiler plate response with no real solution. They have moved a number of my pipelines to the new Engine, and I see marginal or no improvements. Telling developers to try to build more efficient Pipelines is great, but when a single 'if' statement takes 2-5 minutes to process, it doesn't matter how efficient the pipeline is, it will be so slow it is unusable.
I could understand if the pipelines were previously taking 5 seconds, and are now taking 45 seconds. The solutions you propose would be fine. But that is not the case. Pipelines that previously took 5 seconds are taking in some cases over an hour!!!
Even if you achieved 5 times faster that is still 12 minutes+ for one run!
You are telling us Pipelines has grown too fast for you to keep up with, yet also are forcing everyone to move their automations to Pipelines.
If other developers are reading this and share my frustration, does anyone have ideas of what we can do to combat this?
------------------------------
Mike Tamoush
------------------------------
This seems like a boiler plate response with no real solution. They have moved a number of my pipelines to the new Engine, and I see marginal or no improvements. Telling developers to try to build more efficient Pipelines is great, but when a single 'if' statement takes 2-5 minutes to process, it doesn't matter how efficient the pipeline is, it will be so slow it is unusable.
I could understand if the pipelines were previously taking 5 seconds, and are now taking 45 seconds. The solutions you propose would be fine. But that is not the case. Pipelines that previously took 5 seconds are taking in some cases over an hour!!!
Even if you achieved 5 times faster that is still 12 minutes+ for one run!
You are telling us Pipelines has grown too fast for you to keep up with, yet also are forcing everyone to move their automations to Pipelines.
If other developers are reading this and share my frustration, does anyone have ideas of what we can do to combat this?
------------------------------
Mike Tamoush
------------------------------
EvanMartinez
4 years agoQuickbase Staff
Hi Mike,
I'm sorry that you are still feeling frustrated and that is absolutely understandable given the slowness you are experiencing. You are correct that in some instances there are Pipelines still experiencing slowness right now and that is something that needs to be adresses. I'm hoping it might help if I clarify a bit on the engine and why it might help a little right now but will help more in the long term. It actually would have the advantage long term of not just speeding up individual Pipelines so that they themselves run faster but we are also working to address how many can run concurrently.
You are probably familiar with the concept that Quickbase apps often exist in server with several other applications and on occasion an application can feel pain when a lot of requests get lined up and have to be processed in order. While the engine that runs Pipelines is more complex than one line it does still have a limit on its queuing and essentially how many Pipelines get attention at once, in the case of our growing environment and high traffic that can result in slowdowns like we are seeing this last week. We aren't just going to get you through the lines faster by speeding up your individual Pipelines we are also working on opening up more lanes so that you feel less impact due to traffic. So the engine itself will help things run more quickly, best practices will help them be more efficient with that time, and then increased capacity and load balancing improvements will help open up access to resources and more smartly share them where the need arises.
Being able to really focus on this work long term is part of the reason for the Automations end of life program, trying to make both platforms meet their growing demands would have meant not being able to commit more to one without bigger sacrifices in customer experience. Before the announcement since Automations was very familiar to builders its usage was also growing alongside Pipelines meaning both platforms would have needed this kind of performance attention to avoid slowdown and other issues.
It is important for us to hear about your frustrations and experiences though to help push those priorities and innovations. When we hear this dissatisfaction we want to make sure we are communicating what we are doing about it even, even when parts of it are more long term.
------------------------------
Evan Martinez
Community Marketing Manager
Quickbase
------------------------------
I'm sorry that you are still feeling frustrated and that is absolutely understandable given the slowness you are experiencing. You are correct that in some instances there are Pipelines still experiencing slowness right now and that is something that needs to be adresses. I'm hoping it might help if I clarify a bit on the engine and why it might help a little right now but will help more in the long term. It actually would have the advantage long term of not just speeding up individual Pipelines so that they themselves run faster but we are also working to address how many can run concurrently.
You are probably familiar with the concept that Quickbase apps often exist in server with several other applications and on occasion an application can feel pain when a lot of requests get lined up and have to be processed in order. While the engine that runs Pipelines is more complex than one line it does still have a limit on its queuing and essentially how many Pipelines get attention at once, in the case of our growing environment and high traffic that can result in slowdowns like we are seeing this last week. We aren't just going to get you through the lines faster by speeding up your individual Pipelines we are also working on opening up more lanes so that you feel less impact due to traffic. So the engine itself will help things run more quickly, best practices will help them be more efficient with that time, and then increased capacity and load balancing improvements will help open up access to resources and more smartly share them where the need arises.
Being able to really focus on this work long term is part of the reason for the Automations end of life program, trying to make both platforms meet their growing demands would have meant not being able to commit more to one without bigger sacrifices in customer experience. Before the announcement since Automations was very familiar to builders its usage was also growing alongside Pipelines meaning both platforms would have needed this kind of performance attention to avoid slowdown and other issues.
It is important for us to hear about your frustrations and experiences though to help push those priorities and innovations. When we hear this dissatisfaction we want to make sure we are communicating what we are doing about it even, even when parts of it are more long term.
------------------------------
Evan Martinez
Community Marketing Manager
Quickbase
------------------------------
- BusinessIT4 years agoQrew MemberHi Evan,
Thanks for the detailed update and explanation.
For my pipeline, the move to the new engine resulted in great performance. I am seeing it run in 12-15 seconds which is fantastic. Thanks for the continued focus on performance as the usage grows.
Robert
------------------------------
Robert
------------------------------ - MichaelTamoush4 years agoQrew CaptainEvan,
Thank you fr the detailed response. I understand the limitation and the need to open more lanes for true speed increase. I think the main frustration is - if you don't have enough lanes to support everyone on Pipelines, then it does not seem logical to force everyone off automation and into Pipelines. It would seem to me you would want to first, achieve enough lanes and speed to support everyone, and THEN tell everyone to move off of Pipelines.
I think we all can imagine that since you told everyone to move their automations, and then 2 weeks later Pipelines is essentially down, that perhaps too many people migrated over.....
------------------------------
Mike Tamoush
------------------------------- MichaelTamoush4 years agoQrew CaptainWell, after 3 weeks of headaches, frustration, and being told essentially 'there is nothing we can do', boom, this morning everything is running at typical speeds.
No clue what happened..?! Hopefully this isn't a fluke and it doesn't slow down again tomorrow.
Curious if platform wide it is better? Anyone else still experiencing issues?
------------------------------
Mike Tamoush
------------------------------- JeremyAnson4 years agoQrew CadetI'm not sure if this is platform wide, but I've just watched 3 actions complete in a single pipeline in 30 minutes. Is the best next step to raise a customer support case?
------------------------------
Jeremy Anson
------------------------------