RSS is "resident set size"- as I recall from the days when I compiled my own kernels, it's based on a lazily-maintained not-guaranteed-to-be-accurate count of physical memory pages "in use" by the process. On linux, this number may overstate memory use by 50% or more for non-JVM processes. For JVM processes the "overcount" may be much greater.
On Sat, Jun 6, 2009 at 8:46 AM, Jarkko Oranen<chous...@gmail.com> wrote: > >> >> The problem was that it was not as fast as I expected it should be >> given that it was using no less than 100% of the CPU on my system. >> (two 3GHz Xeon CPUs [1 core]; 3GB RAM; a beefy workstation). That >> this was visible in the GUI shows how slow it appeared to me. Also, >> it was using 700MB RAM (VIRT in top). Sure - it was "swapped" (I'm >> familiar w/ some of the interpretations of these memory issues) except >> that my system has ZERO swap space. PMAP showed that 400MB of it was >> heap, too, not libraries or binaries or anything else that we can >> safely ignore. This was (apparently) real, allocated heap, private to >> the process's address space. > > I doubt the VIRT size matters at all. From what I know, it represents > the address space that is available to the process; there's no > guarantee that it's actually allocated or in use. Hence, it's not even > "swapped". On my system, I have several processes with hundreds of > megabytes in their "VSIZE" column shown in top. About 12GB in total; > yet, I have barely any swap space... only a single 64MB file is > allocated. (OS X allocates swapspace dynamically) > According to top, I have 0 pageouts since last boot, so the swap isn't > even being used. I have 3GB of RAM. > >> Additionally, as the simulation ran, the initial RSS of 60MB rose to >> 130MB then stopped. The VIRT remained constant. I had expected that >> - however I remained concerned. > > I'm not sure what kind of memory "RSS" is, but if it's shared, then > it's java's own overhead, and not the application itself. > the "private" memory areas are what you're interested in. Though of > course the overhead is meaningful if it's the only java app running, > but it's very difficult to tell what amount of the shared memory is > actually used only by your application. > > You can try tuning the java VM and decrease its heap size and other > things; see if it makes a difference. > > -- > Jarkko > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en -~----------~----~----~----~------~----~------~--~---
