-
Notifications
You must be signed in to change notification settings - Fork 1
Open
Description
A very common way to call the BMI methods is
while m.get_current_time() < m.get_end_time():
m.update()
for v in variables:
output[m.get_current_time()][v] = m.get_value(v)Lots of requests are needed to get a value after each update, this could be slow.
To speed up running a model it would be nice to have additional run functions like
update_until_and_get_value_of_variables(time, variables): Mapping<str, list<float>>update_until_and_get_value(time, variables): list<float>>update_until_and_get_value_at_indices(time, variable, indices): list<float>update_until_and_get_value_of_variables_at_indices(time, variables, indices): Mapping<str, list<float>>
We should benchmark to see if this how slow it is, if slow we should implement those functions.
The responses can get quite big, the client and server should support this.
Could those functions be a BMI extension, see csdms/bmi#138 ?
Similar to eWaterCycle/grpc4bmi#152
Metadata
Metadata
Assignees
Labels
No labels