11-10-2010
There's no hard rule on this. Some people over commit.. so the sum total of VM resources is greater than actual physical resources. Usually, one place where this is done is with regards to CPUs. BUT... if your VMs REALLY need a whole bunch of CPU resources, it's possible that you'll reach the limits of what you can actually give and have a "good" performing platform. IMHO, it's pretty safe to say that we waste about 50% of CPU resources (in general, the number is usually much, much higher). So you could use that for a very conservative number of total CPUs you can dole out. And don't discount the hyperthreads... they do help out as well...
If your plan is to virtualize large scale servers (e.g. 4 CPUs + 8G+ memory) ... I think that's a BAD plan. Virtualization is most effective when talking about smaller footprint machines. Then the scalability can be VERY large.. and you can really over commit (but again, there is no RULE).
10 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
Is there an easy way in Solaris to count the number of processors? (2 Replies)
Discussion started by: hshapiro
2 Replies
2. UNIX for Dummies Questions & Answers
hi,
i want to know cpu utilizatiion per process per cpu..for single processor also if multicore in linux ..to use these values in shell script to kill processes exceeding cpu utilization.ps (pcpu) command does not give exact values..top does not give persistant values..psstat,vmstat..does njot... (3 Replies)
Discussion started by: pankajd
3 Replies
3. Solaris
Hello Friends,
On one of my Solaris 10 box, CPU usage shows 100% using "sar", "vmstat". However, it has 4 CPUs and prstat and glance are not showing enough processes to justify high CPU utilization.
=========================================================================
$ prstat -a
... (4 Replies)
Discussion started by: mahive
4 Replies
4. AIX
I have a number of LPARs on one P520. All LPARs are running 5.3 and I observe the following:
On some LPARs the number of CPUs found do not match between topas and mpstat.
Server 1:
$ mpstat
System configuration: lcpu=4 ent=0.2 mode=Uncapped
cpu min maj mpc int cs ics rq ... (1 Reply)
Discussion started by: petervg
1 Replies
5. Red Hat
Hi,
I am observing high load average on RHEL5 server and need to check the number of core processors available on the box to give me a bigger picture on whether or not I should be worried.
I have always checked the physical count quite simply.....
# grep 'physical id' /proc/cpuinfo |... (10 Replies)
Discussion started by: Duffs22
10 Replies
6. Red Hat
Hello,
sysinfo throws out below 3 CPU counts. Can anyone help me understand what each of these means?
CPU Count Socketed is 2
CPU Count Physical is 8
CPU Count Virtual is 16
First one seems obvious. However, I wonder how there can be 8 Physical CPUs, if... (2 Replies)
Discussion started by: hnhegde
2 Replies
7. HP-UX
Hello,
I am trying to figure out the maximum number of 11.31 IVMs we can create on our system. We have an Integrity rx2660, and here is the hardware info:
CPU info:
2 Intel(R) Itanium 2 9100 series processors (1.42 GHz, 12 MB)
532 MT/s bus, CPU version A1
4 logical... (2 Replies)
Discussion started by: bstring
2 Replies
8. UNIX for Dummies Questions & Answers
We have a single threaded application which is restricted by CPU usage even though there are multiple CPUs on the server, hence leading to significant performance issues. Is it possible to merge / combine multiple CPUs at OS level so it appear as a single CPU for the application? (6 Replies)
Discussion started by: Dissa
6 Replies
9. Solaris
Hi All,
Currently experiencing more-so a visual issue when typing psrinfo, mpstat commands, as the virtual processors start from 8-39, as below:
# psrinfo -pv
The physical processor has 32 virtual processors (8-39)
UltraSPARC-T2 (chipid 0, clock 1165 MHz)
Can't seem to find anything to... (0 Replies)
Discussion started by: alx_mck
0 Replies
10. AIX
seeing weirdness across some 7.1.5.1 LPARs - they all have 2 vCPUs allocated shared. With prtconf they show 2 CPUs, 'lsdev -c processor' concurs, and 'lsattr -El procX' shows that SMT is enabled and there are 2 SMT threads (power5, sorry). Yet running topas on them shows 2 CPUs on some and 4 on... (3 Replies)
Discussion started by: maraixadm
3 Replies
LEARN ABOUT SUSE
mtrr_add
MTRR_ADD(9) Hardware Interfaces MTRR_ADD(9)
NAME
mtrr_add - Add a memory type region
SYNOPSIS
int mtrr_add(unsigned long base, unsigned long size, unsigned int type, bool increment);
ARGUMENTS
base
Physical base address of region
size
Physical size of region
type
Type of MTRR desired
increment
If this is true do usage counting on the region
DESCRIPTION
Memory type region registers control the caching on newer Intel and non Intel processors. This function allows drivers to request an MTRR
is added. The details and hardware specifics of each processor's implementation are hidden from the caller, but nevertheless the caller
should expect to need to provide a power of two size on an equivalent power of two boundary.
If the region cannot be added either because all regions are in use or the CPU cannot support it a negative value is returned. On success
the register number for this entry is returned, but should be treated as a cookie only.
On a multiprocessor machine the changes are made to all processors. This is required on x86 by the Intel processors.
The available types are
MTRR_TYPE_UNCACHABLE - No caching
MTRR_TYPE_WRBACK - Write data back in bursts whenever
MTRR_TYPE_WRCOMB - Write data back soon but allow bursts
MTRR_TYPE_WRTHROUGH - Cache reads but not writes
BUGS
Needs a quiet flag for the cases where drivers do not mind failures and do not wish system log messages to be sent.
COPYRIGHT
Kernel Hackers Manual 2.6. July 2010 MTRR_ADD(9)