"Group by" in Hive refers to dividing each data set into several small data sets through certain rules, and then grouping data for several small data sets

Which of the following options are the key features of Zookeeper? (Multiple choice)

Assuming that the amount of data is about 200GB, and the maximum capacity of shards is limited to 30GB, what is the maximum number of shards that should be designed?

Flink has no restart strategy.

In the Fusioninsight Hadoop cluster, through df-hT query on a certain node, the partitions seen include the following:
/var/log
/srv/BigData
/srv/BigData/hadoop/data5
/srv/BigData/solr/solrserver3
/srv/BigData/dbdataom
What is the planning combination of the optimal Raid level for the disks corresponding to these partitions?