Apache Spark has been a popular choice for large-scale distributed data processing. However, as data teams move to cloud architectures and separate computes from client interfaces, the traditional tightly coupled Spark driver model has begun to revea...