I am implementing a new algorithm for scheduling in hadoop called
TaskTrackerAware Scheduler. I have to configure some properties such as mapred.tascheduler.task.max ( maximum number of tasks that can run on a tasktracker for a single job) and mapred.tascheduler.hosts (host names of the task tracker in which jobs need to be run). How to configure these properties in Job configuration object?
Find below the snippet that may help you,
Configurarion conf = new Configuration(); conf.set("mapred.tascheduler.task.max" , <value>); conf.set("mapred.tascheduler.hosts " , <value>); Job job = Job.getInstance(conf, "app name");
Let me know if you need any further help..