java - HashMap as a Broadcast Variable in Spark Streaming? -

i have data needs classified in spark streaming. classification key-values loaded @ beginning of program in hashmap. hence each incoming data packet needs compared against these keys , tagged accordingly.

i realize spark has variables called broadcast variables , accumalators distribute objects. examples in tutorials using simple variables etc.

how can share hashmap on spark workers using hashmap. alternatively, there better way this?

i coding spark streaming application in java.

in spark can broadcast serializable object same way. best way because shipping data once worker , can use in of tasks.

scala:

val br = ssc.sparkcontext.broadcast(map(1 -> 2)) 

java:

broadcast<hashmap<string, string>> br = ssc.sparkcontext().broadcast(new hashmap<>());