Process "Killed" Loading Large CSV in Binderhub

[psychemedia]

Trying to load a 1Gb / 2.5 million rows CSV file into a table on resource limited MyBinder VM, I kept getting a Killed message on the process.

What did work fine was a chunked load using pandas, although without predefining the table is does mean the table structure could end up being anything...:

for chunk in pd.read_csv(datafile, chunksize=2000):
    chunk.to_sql(name=tablename, con=conn, if_exists="append", index=False)