Hi Dennis,

You might not have enough RAM for 1,000. (Only 2MB/zone! Not counting 
RAM needed for the kernel) For discussion, see Jeff Victor's blog 
http://blogs.sun.com/JeffV/ especially the 3 entries titled "Spawning 
0.5kZ/hr". He reasched 1,000 zones, but he had more RAM.  There are a 
few kernel tunables he mentions that you'll need to adjust.  I suggest 
bring up an initial zone, turning off services you won't use in your 
test (eg: svcadm disable webconsole; svcadm disable sendmail) to reduce 
the memory footprint and number of processses.

Each will have a JRE, even within a sparse root - No extra effort for 
that.  Each zone will have its own IP address too. I'll let somebody who 
knows the other networking parts answer them

regards, Jeff S

Dennis wrote:

>a friend of mine has written a P2P application in Java as a master thesis. Now 
>we want to setup a testing configuration using containers to simulate the 
>function of each node. Therefore I have several question, any help is 
>- We are planning to test 1000 nodes (=containers) on a 2 GB RAM Athlon Dual 
>Core 4.2 GHz machine. Is that possible?
>- Each container should be a sparse zone that contains no application but a 
>JRE. How do I create one of these?
>- Each node has its own IP adress
>- To simulate it as realistic as possible, we want to give each node a certain 
>bandwith with crossbow. Is that possible with the lo interface or only with an 
>external nic?
>- What is the recommended way to examine the traffic on a vnic if possible, or 
>on each container?
>- Is there a way to check if a certain file is read or written in a container 
>directory from the global zone?
>Thanks for your help!
>This message posted from opensolaris.org
>zones-discuss mailing list

Jeff Savit
Principal Engineer
Sun Microsystems, Inc.        Phone:  732-537-3451 (x63451)
400 Atrium Drive              Email:  [EMAIL PROTECTED]
Somerset, NJ  08873           http://blogs.sun.com/jsavit/

zones-discuss mailing list

Reply via email to