Central model up-/downtime management

Central model up-/downtime management

After @DaanishSoomar posted this article , I wondered why there is actually no central admin setting to wake-up a model automatically.

Most customers I know with larger models have either hourly imports during working hours or they run specific wake-up scripts.

Particularly larger, complex models can take 10-15 minutes until they are open (or the standard 1-2 coffees if you are the first one in the office).

 

As a model- / workspace admin I want to be able to set central "up-/downtimes" for each model so that there is no need for script based wake-up calls.

During up-time the model would always remain in core. During down-time it would only load into the core on demand. 

6 Comments
alexpavel
Certified Master Anaplanner

As a workaround, it is also the possibility to create an integration with CloudWorks which has schedule capabilities.

It can be created a dummy file on Amazon S3 and launch an integration into a dummy model that needs to be wakened up. 

 

I think also that there is some timeout that if a model is not used for some time, it goes back to "sleep". 

 

 

 

PhilippErkinger
Certified Master Anaplanner

@alexpavel But this would require an AWS license, if I'm not mistaken?

alexpavel
Certified Master Anaplanner

For limited usage, there is AWS free tier..but indeed currently this needs to be taken into consideration.

 

My personal opinion: CloudWorks should be open also for launching processes for integration inter-models without to be obliged to have a source from AWS.

 

below an idea about this:

 

https://community.anaplan.com/t5/Idea-Exchange/CloudWorks-Possiblity-to-launch-processes-without-bei...

 

 

 

PhilippErkinger
Certified Master Anaplanner

Absolutely! Great idea

johan_vangerwen
Frequent Contributor

As with Cloudworks you may consider the Model to Model functionality, no files no AWS license needed.

angs75
Certified Master Anaplanner

This is a great idea and we can use CloudWorks to call a blank process as @DaanishSoomar used in his Article. in that case you don't need any AWS. We are planning to use it for All large production modules and in flight UAT modules. in that case User Experience would be much better.

Do anybody know a model goes to dormant stage after how many minutes?