Dmitri Trembovetski wrote:
> Hi Garrett,
>
> > be much, much smaller. And I doubt the problem is the Java code. Even
> > JSpider on my AMD64 Ultra20 is 94 MB. The JVM is a pig.
>
> Take a look at this blog:
>
> http://weblogs.java.net/blog/xiaobinlu/archive/2005/08/perception_real.html
>
> In Java 6 the (perceived) footprint you're seeing should be reduced.
>
> Thanks,
> Dmitri
> Java2D Team
>
That's good. I didn't realize Java was keeping the entire JAR library
mmap()'d.
It strikes me that this problem underscores that we need a better way to
monitor real memory usage -- perhaps looking at memory allocated via
brk() vs memory backed by mmap() of files (whether ordinary as in the
jar or special as in framebuffers.)
-- Garrett
> On Mon, Jun 26, 2006 at 12:13:15PM -0700, Garrett D'Amore wrote:
> > Dan Davies Brackett wrote:
> > > (crossposting the reply too.)
> > >
> > > Cacao is the Common Agent Container, the single JMX container that all
> > > Sun's manageability solutions are supposed to run in. I know it gets
> > > used for basic system information collection for Software Update and
> > > the registration nagware that comes up when you log in, which brings
> > > me to the next line...
> > > swupna.jar is probably related to Software Update; have you tried
> > > registering (it's free) for software update and letting it get its
> > > little icon out of the JDS tray?
> >
> > Yes, I did register using the nagware.
> >
> > I'm going to reboot this system and see if it doesn't start those Java
> > apps. It really is quite unfortunate that these things are piggy as
> > they are. This is a strong argument against writing "standard" system
> > software in Java, IMO -- the equivalent C functionality would probably
> > be much, much smaller. And I doubt the problem is the Java code. Even
> > JSpider on my AMD64 Ultra20 is 94 MB. The JVM is a pig.
> >
> > -- Garrett
> > >
> > > Garrett D'Amore wrote:
> > >> I've noticed something scary. My UltraBook IIi has 256 MB. Here's the
> > >> output from top with nothing except JDS running:
> > >>
> > >> last pid: 2314; load averages: 0.08, 0.32, 0.36
> > >> 11:13:42
> > >> 67 processes: 62 sleeping, 4 running, 1 on cpu
> > >> CPU states: 97.4% idle, 0.7% user, 2.0% kernel, 0.0% iowait, 0.0%
> > >> swap
> > >> Memory: 256M real, 54M free, 228M swap in use, 435M swap free
> > >>
> > >> PID USERNAME LWP PRI NICE SIZE RES STATE TIME CPU COMMAND
> > >> 2314 root 1 59 0 2072K 1632K cpu 0:00 0.56% top
> > >> 1138 root 22 59 0 115M 42M run 0:37 0.49% java
> > >> 1370 root 14 49 0 105M 18M run 0:06 0.35% java
> > >>
> > >> Notice that the two Java process are sucking up a large chunk of
> > >> memory. Even if you only look at resident memory, these occupy well
> > >> over 50MB.
> > >>
> > >> What are these two Java processes? Well the "big" one (42MB) appears to
> > >> be running com.sun.cacao.container.impl.ContainerPrivate, and the
> > >> smaller (18M) one appears to be running /usr/lib/patch/swupna.jar
> > >> (-wait).
> > >>
> > >> Again, this is UltraSPARC platform stuff. Admittedly 256 MB is a little
> > >> smallish for RAM these days, but still....
> > >>
> > >> What the heck is Cacao? And why does Java want so much memory? I
> > >> should probably ask this on one of the other lists than laptop-discuss,
> > >> but since laptops are often more resource constrained, it still seems
> > >> somewhat appropriate to raise the issue here.
> > >>
> > >>
> > > _______________________________________________
> > > laptop-discuss mailing list
> > > laptop-discuss at opensolaris.org
> >
> >
> > --
> > Garrett D'Amore, Principal Software Engineer
> > Tadpole Computer / Computing Technologies Division,
> > General Dynamics C4 Systems
> > http://www.tadpolecomputer.com/
> > Phone: 951 325-2134 Fax: 951 325-2191
> >
> > _______________________________________________
> > desktop-discuss mailing list
> > desktop-discuss at opensolaris.org
>
--
Garrett D'Amore, Principal Software Engineer
Tadpole Computer / Computing Technologies Division,
General Dynamics C4 Systems
http://www.tadpolecomputer.com/
Phone: 951 325-2134 Fax: 951 325-2191
