custom designed profiling

[maha <ma...@umail.ucsb.edu>]

Hello,

  I'm wondering if there is a way similar to org.apache.hadoop.mapred.JobProfile but goes further inside the job.

Example:
	
		Map () { 
			.....
	  java.util.Date d = new Date();
	  dt.getTime(); 
	  Runtime.getRuntime().freeMemory(); 
	  Runtime. getRuntime().totalMemory(); 
	 		....
		}

I want to find the time/memory used by a specific OBJECT or BLOCK of CODE.  

I think it's possible to make mappers write those values into files ... but then I'm wasting time opening a file. 

Is there a faster way to do that such as background loggers saving mappers output ??

Thank you,

Maha