Oracle is running on one of the zone (out of 17 non-global zones) and it is supposed to use 1 core only, but looks like it is not configured correct. Below is the output from global as well as zone
What is significance of limit=1 in this output from zonecfg ?
Only one zone is having limit=2 and rest all 16 zones are having limit=1. Will it not be restrcing cores anbd Oracle should see and charged only for one core ?
No. You will need to create a pset with wanted cpu cores and alocate that to specific zone or zones to compliant as far as i recall.
Lets say for simplicity you have 10 zones, 1 global zone and 9 zones
9 zones have 1 vcpu share (including global), while 1 zone has 2 vcpu share.
This means that in only in the case of cpu pressure, say 100%, kernel will guarantee each zone 10% of resources (1 cpu share) and one zone (2 cpu shares) will get 20% of cpu resources.
Not cores per se, but compute power using FSS scheduler.
For instance, all zones are mostly idle using total of 10% cpu, any zone will able to use rest of cpu time, as long there is no competition for cpu time.
Only under cpu pressure the scheduler will kick in to alocate time per ratio (1:1:1:1:1:1:1:1:1:2 = 100) to ensure quality of service per shares configuration.
When using psets allocated to zones, a zone will have dedicated cores defined in pset and no other zone will able to use those cores.
And you will be compliant as oracle hard partition, if that didn't change in the meantime...
Hope that helps.
Regards
Peasant.
These 2 Users Gave Thanks to Peasant For This Post:
Hi all.
I have a question about linux command to find number of CPU and Core.
I usually use the command dmidecode -t processor to find cpu and core numbers . On this machine with Red Hat 4. 0 when I try to insert the command is returned the error
-bash: dmidecode: command not found
I try to... (8 Replies)
Hi all !
I'm new in this site, so sorry if this question is into wrong place.
How can I limit cpu/core and memory usage by user?
System: RedHat Ent. Linux. 6.4
Tks, (4 Replies)
Hi Gurus
Can someone help me in explaining the below outputs .
psrinfo -p
4
/usr/sbin/psrinfo -pv
The physical processor has 4 virtual processors (0-3)
SPARC64-VI (portid 1024 impl 0x6 ver 0x93 clock 2150 MHz)
The physical processor has 4 virtual processors (8-11)
SPARC64-VI... (3 Replies)
Hi Experts,
I am executing multiple instances(in parallel) of perl script on HP-UX box.
OS is allocating substantial amount of CPU to these perl processes,resulting higher cpu utilization.
Glance always shows perl processes are occupying majority of the CPU resource. It is causing slower... (2 Replies)
Hi folks,
I want to know how to run two unix programs on two different cpu cores on a 2-core or 4-core or 8-core CPU machine? Extending this how would i run four and eight unix programs on 4-core and 8-core machine respectively?
If this can be done, how to know which program is assigned to... (1 Reply)
Hey all,
dmidecode | grep -i CPU
Socket Designation: CPU 0
Version: Intel(R) Xeon(R) CPU E5530 @ 2.40GHz
Socket Designation: CPU 1
Version: Intel(R) Xeon(R) CPU E5530 @ 2.40GHz
cat /proc/cpuinfo | grep -i cpu
cpu family : 6... (24 Replies)
I wrote a very simple script that matches combinations of alphabetic characters (1-5). I want to use it to test CPU speeds of different hardware/platforms. The problem is that on multi-core/processor systems, only one CPU is being utilized to execute the script. Is there a way to change that?... (16 Replies)