scottp at dd.com.au
Wed Sep 13 21:16:18 PDT 2006
On 14/09/2006, at 13:49, Daniel Pittman wrote:
> Scott Penrose <scottp at dd.com.au> writes:
>> On 14/09/2006, at 13:12, Andrew Speer wrote:
>>> On Thu, September 14, 2006 11:35 am, Scott Penrose wrote:
>> My experience with the above is you get the same memory footprint. As
>> I said, it is not mod_perl, it is Perl that is large.
>> I am amused by the 10M process too because our process are normally
>> about 300MB using perl, and Java Tomcat starts at about 800MB+ - we
>> have servers that one process is GB for TomCat - but it still runs at
>> a few hundred hits a second. Although it is slow not because of the
>> process size but because of what it is doing.
> TomCat was, last time I checked, fairly misleading: the JVM was doing
> all sorts of clever mapping tricks to make the garbage collection
> It had a stupid quantity of memory mapped, but almost exclusively
> sparsely used, and so mostly a "virtual" mapping that didn't have any
> real backing, not even a page of zeros.
Maybe all true, but it was eating up swap like CRAZY. Once we
increased the machines ram form 1GB (the size of the JVM) to 1.5GB
and stopped using swap, most things were ok.
Like mod_perl size though, it then does not matter about fork/threads
because as we have said - copy on write.
More information about the Melbourne-pm