It is possible to use MapReduce on heterogeneous environment without any extra interface. If hard disk capacity is different, then extra capacity will be added in hdfs and if more memory is present then depending on the task requirement it will be used.
No. of slots per task tracker cannot be varied so even if some nodes have additional cores, extra slots cannot be added. Regards, Jayant -----Original Message----- From: Sujitha [mailto:gs...@tce.edu] Sent: Thursday, March 11, 2010 1:09 PM To: common-dev@hadoop.apache.org Subject: Map Reduce in heterogeneous environ.. hi all map reduce is running on hadoop clusters(homogeneous).is it possible to use it in a heterogeneous environment..is that we need any interface to support this?? suggestion please. Regards Suji ----------------------------------------- This email was sent using TCEMail Service. Thiagarajar College of Engineering Madurai-625 015, India
