Skip to content


Runs a future computation and returns its result as an actual Julia value

compute(future; destroy = [])


Banyan is a suite of libraries with functions for processing big data. These functions operate on and return futures. To convert a future into an actual Julia value, you may either compute it for an exact result or sample it for an approximation.

Required Parameters

future (Future)

The future to compute.

destroy (Vector{Future})

Futures to mark as destroyed so that they don't get saved to disk (saving to disk can be very time-consuming).

Returns (Any)

The Julia value that results from computing the future.

Example Usage

df = read_parquet("s3://inventory/availability.parquet")
df_filtered = filter(row -> row.availability < 3, df)