I now work on the hop in pseudo-distributed mode. I try some Madrodas, package it as a jar, and copy the file to the fancy. Then use
./bin / hadoop jar * to start it . My question is: is there any another way to do this ? If we have thousands of jobs to run, we can not just type in command. What do we do in the production environment? Thank you. If you have 1000s of jobs, then write a shell script and submit them if the jobs are between If there are dependencies then try to use Apache Ozie as Chris mentioned it.
Comments
Post a Comment