Tasks running out of memory and mapred.child.ulimit

[Steve Lewis]

I have a small hadoop task which is running out of memory on a colleague's
cluster.
I looked at has mapred-site.xml and find

  <property>
    <name>mapred.child.java.opts</name>
    <value>-Xmx512M</value>
  </property>
  <property>
    <name>mapred.child.ulimit</name>
    <value>1572864</value>
  </property>

I set mapred.child.java.opts  <value>-Xmx1024M</value>

but in my cluster there is no value for mapred.child.ulimit
I do not usderstand this parameter and wonder
    what the default is?
    what value is appropriate for -Xmx1024M?
   what value is appropriate for -Xmx512M?

Are there other parameters I should look at

-- 
Steven M. Lewis PhD
4221 105th Ave NE
Kirkland, WA 98033
206-384-1340 (cell)
Skype lordjoe_com