Fixed Fan-In with Order-Only Dependencies I'm trying to create a graph with data and order-based dependencies (see attached diagram).

op A

produces some data that ops in

graph C

depend on (data dependency).

op B1

,

op B2

,

op B3

, all need to complete before

graph C

should start (order-based dependency).

op B1

,

op B2

,

op B3

execution order is not important. Can anyone suggest a way to build this graph please?

something like

@graph
def example():
    a = op_a()
    b1 = op_b1(a)
    b2 = op_b2(a)
    b3 = op_b3(a)
    graph_c(data=a, start_after=[b1, b2, b3])

should work, though if

graph_c

is a

@graph

, you will need to map the input through to all the inner ops that should block / receive data

start_after

? What is this magick?

just me extrapolating an example, it assumes you map

start_after

in

graph_c

to some op that takes something like

ins={'start_after': In(Nothing)}

@alex 😄 OK, so not magick then, just order-based dependency. I had tried what you suggest here, but ran into the problem you warn against, namely needing to map all the `start_after`™️ inputs to ops in the graph. This very quickly turned into a schmozzle and I was hoping for a cleaner and less fragile pattern.

feel free to chime in here https://github.com/dagster-io/dagster/issues/4274