- how map tasks are distributed among nodes.
- Can i set nodes to run in parallel. for example if datanodes are 2, and map tasks are 4 which of them take the task? i see from the manager sometimes datanode takes the task and sometimes the other datanode works. The job is given to both nodes and when any of them is assigned a task it runs one at a time. on what basis do the namenode assign tasks.
- Is it done manually in the code or done auto. If there is a link to a tutorial on how tasks run on nodes it will be great if you provided it.
Asked
Active
Viewed 104 times
1
std
- 13
- 4