hadoop-mapreduce-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jyothish Soman <jyothish.so...@gmail.com>
Subject In Memory Filesystem
Date Sun, 20 Jun 2010 08:18:24 GMT
Hello again,

I ran into another issue with pseudo distributed mode, I allocated 10GB for
in memory filesystem. I ran sort (not terasort) with 10 GB of data, and
there was absolutely no effect of the in memory fs. The number of writes did
not reduce in anyways from the map phase, any suggestions?.
Is it the intrinsic behaviour, or did I miss something.

Jyothish Soman

View raw message