What do I loose if I run spark without using HDFS or Zookeper ? which of them is
almost a must in practice?

Reply via email to