when i run...how do I define the run order of flow zones

elitrn25

We have 8 flow zones, everything works by itself but we didn't test all the flow zones together (the flow zones are connected by some datasets) ... I need to know how to define the order of building/run of the flow zones or at least I should know how it works.

Can you help me giving me some more info about it?

Zach

Hi @elitrn25
,

Flow zones don't affect the order that datasets are built. The build order is determined based on how datasets are connected to each other. For example, if datasetA is an input for datasetB, then datasetA will be built before datasetB even if they're in different flow zones.

You can read more about building multiple datasets here: Concept: Dataset Building Strategies.

Thanks,

Zach

tgb417

@elitrn25
,

Although @ZachM
is correct when using DSS in the standard way. If you want to control rebuild more specifically and have access to a full Dataiku License, you can use scenarios to change the build order of data sets by explicitly calling out which datasets you want to build. There is also an advanced feature on datasets that can stop how far back a flow goes when you ask for a recursive rebuild. That is the option require explicit rebuilds. This means that if you choose a dataset late in the flow and you force a recursive rebuild the rebuild will only go back as far as the item in the flow are set to normal. If you make the last dataset in each flow zone explicit only. You may be able to do what you are thinking of doing. With a combination of these two features I find that I can control which flow zones or set of flow zones I want to rebuild.

Hope that might help.