1 Resposta
- Mais recentes
- Mais votos
- Mais comentários
0
Not really, you would need a distributed transactions coordination between nodes that neither Glue nor Spark support.
What you can do is instead of writing to the final table, write to a temporary table and once the write operation is complete issue a DB command (using JDBC or other library) to do the merge inside the DB in a transactional way.
Conteúdo relevante
- AWS OFICIALAtualizada há 7 meses
- AWS OFICIALAtualizada há 2 anos