On 05/09/2006, at 11:39 PM, Nergal Dimitri wrote:
I got a few questions;

1.) Is the 'priocntl' utility used instead of the 'nice' utility, or is it just another priority tool?
      I got the feeling it's the same as 'nice'....

I can't remember where I saw this documented, but basically priocntl is the preferred and native method of changing scheduling parameters on solaris. The nice command is there for compatibility reasons but provides far less control than priocntl. For example priocntl allow selection of the scheduling class and class-specific parameters.

2.) Is Dynamic System Domains the same as zones? The description of it feels like it's _almost_ the same, but more hardware oriented division. (Link for DSD information?)

Not the same. DSDs (more commonly known simply as domains) are the result of hardware level partitioning in the mid-rtange and high end machines. Each runs it's own instance of an OS, potentially different versions and has dedicated and isolated hardware. Zones are a software partitioning of a single OS instance that requires no hardware support but provides less in the way of OS-version and harware-fault isolation. You can, of course run zones on a domain.

Argh.. sorry I've just been called away, so I'll have to leave your other questions for others, or for later.

Boyd.

3.) What's the actual difference between changing user priority and global priority? What happends if I change both to maximum (say RT). Is the user priority less worth the global? If I as root change the user priority, is there any point in changing the global priority of the process/project?

4.) If I set a process/project to have a certain limit of memory allocation, could this have positive effects on applications with memory leaks? If the memory leaks cannot pass say 50Mb of memory, can these 50Mb be allocated with non-usable memory (memory leaks) so the actual memory usage gets even worse for the application? And if I have have 4Gb of memory, and I set a limit of 3.5Gb to one application that at a certain point only use 2Gb of the memory and another application uses 2Gb. Then suddenly the first application (the one with 3.5Gb limit) uses its full limit (3.5Gb), what happends with the memory that the second application had allocated (1.5Gb)? Does this memory get swapped to the HDD and the second application only get access to the 500Mb memory thats left?


I hope you understand my questions even if they got a bit sloppy :)

Best Regards,
Nergal


This message posted from opensolaris.org
_______________________________________________
opensolaris-discuss mailing list
[email protected]

_______________________________________________
opensolaris-discuss mailing list
[email protected]

Reply via email to