Will a spark deploy on neo4j side accelerate my writing?

Neo4j Graph Platform Import / Export
1

I am writing data from my hdfs server into my neo4j server using pyspark.

df.write.format("org.neo4j.spark.DataSource").mode("Overwrite").option("url", 'neo4j://192.xx.xx.xxx1:7687') \
    .option("authentication.basic.username", "username").option("authentication.basic.password", "password") \
    .option("query", "match (n:entity {xnamex:'john doe'}) with n set n.xlabelx='temp'").save()

Yet I realized spark (or pyspark) is not installed on my neo4j server, while the code still succeeded.

I just wonder that, do I only need neo4j-spark-connector on my datasource side? Or I can also deploy spark on neo4j side so it will lead to an acceleration？