SEO

Hi,

Is it possible to create two partition on IBM p630 ?
Partition 1 with SCSI 1 / LAN 1
Partition 2 with SCSI controler #2 and lan #2 ?

regards,
Bruno

Yes, but I believe the hd's cannot be on the same scsi bus.
Alok
"bruno Monnet" <monnetb@yahoo.com> wrote in message
news:2a0f4cc4.0406080748.d5f1b9d@posting.google.co m...
> Hi,
>
> Is it possible to create two partition on IBM p630 ?
> Partition 1 with SCSI 1 / LAN 1
> Partition 2 with SCSI controler #2 and lan #2 ?
>
> regards,
> Bruno

"bruno Monnet" <monnetb@yahoo.com> schrieb im Newsbeitrag
news:2a0f4cc4.0406080748.d5f1b9d@posting.google.co m...
> Hi,
>
> Is it possible to create two partition on IBM p630 ?
> Partition 1 with SCSI 1 / LAN 1
> Partition 2 with SCSI controler #2 and lan #2 ?
>
> regards,
> Bruno

Yes.
You cannot share devices (yet) so you need at least one separate
controler/disk per partition and a separate processor of course. One
drawback I found with such small servers: should you happen to have 4 GB of
RAM expect to loose 756MB for memory handling. (original quote from the IBM
support hotline: "hypervisor needs 256MB and TCE needs 256 MB - this equals
756 MB")

Regards,
Andreas ;-)

> Is it possible to create two partition on IBM p630 ?
> Partition 1 with SCSI 1 / LAN 1
> Partition 2 with SCSI controler #2 and lan #2 ?
>

If you have the split backplane feature. In that case the disk bay will
be split in 2 allowing for 2 partitions on the base p630. Another option
is to have boot from FC or to have an additional I/O Drawer.

BC

Bernardo Cabral wrote:
>> Is it possible to create two partition on IBM p630 ? Partition 1 with
>> SCSI 1 / LAN 1 Partition 2 with SCSI controler #2 and lan #2 ?
>>
>
> If you have the split backplane feature. In that case the disk bay will
> be split in 2 allowing for 2 partitions on the base p630. Another option
> is to have boot from FC or to have an additional I/O Drawer.
>
> BC

No, that is a feature of the p650. The 630 is
a single 4-pack in the base CEC unit.

The best synopsys of 630 LPAR is the 630 Technology
and Architecture paper and the Complete Partitioning
Guide, both available at www.redbooks.ibm.com.

Greg.

>
> No, that is a feature of the p650. The 630 is
> a single 4-pack in the base CEC unit.


I think since then the p630 was updated to get the split backplane.

BC

The Hypervisor and TCE are a minimum of 256MB each (TCE can be bigger on
large machines like the p690 with many remote I/O drawers) and are like a
Logical Partition tax.
The other 512MB to 756MB bit is used for the Page Tables used to manage
virtual memory (they are virtual address to physical address translation
tables used by the CPU H/W).
After an initial chunk, this is 1/64th of memory.
This can be seen in LPAR machines because the Hypervisor is managing virtual
memory to ensure the LPARs can never share the same physical memory (LPARs
are isolated) abd is seen on one of the HMC machine proprty panels.
Exactly, the same amount of memory is used when not in LPAR mode - i.e. SMP
mode or just running on a normal RS/6000 or pSeries machine but you are
never told how much memory is used by these same Page Tables.
So yes the 512MB is an overhead you pay for LPAR but the rest is not.

Hope this helps, Nigel Griffiths
>
> Yes.
> You cannot share devices (yet) so you need at least one separate
> controler/disk per partition and a separate processor of course. One
> drawback I found with such small servers: should you happen to have 4 GB
of
> RAM expect to loose 756MB for memory handling. (original quote from the
IBM
> support hotline: "hypervisor needs 256MB and TCE needs 256 MB - this
equals
> 756 MB")
>
> Regards,
> Andreas ;-)
>
>

"Nigel Griffiths" <nag@uk.ibm.com> schrieb im Newsbeitrag
news:caefuo$juc$1@hercules.btinternet.com...
> The Hypervisor and TCE are a minimum of 256MB each (TCE can be bigger on
> large machines like the p690 with many remote I/O drawers) and are like a
> Logical Partition tax.
> The other 512MB to 756MB bit is used for the Page Tables used to manage
> virtual memory (they are virtual address to physical address translation
> tables used by the CPU H/W).
> After an initial chunk, this is 1/64th of memory.
> This can be seen in LPAR machines because the Hypervisor is managing
virtual
> memory to ensure the LPARs can never share the same physical memory (LPARs
> are isolated) abd is seen on one of the HMC machine proprty panels.
> Exactly, the same amount of memory is used when not in LPAR mode - i.e.
SMP
> mode or just running on a normal RS/6000 or pSeries machine but you are
> never told how much memory is used by these same Page Tables.
> So yes the 512MB is an overhead you pay for LPAR but the rest is not.
>
> Hope this helps, Nigel Griffiths
> >

Hi Nigel,

thanx for the theorie ;-) However the example I gave is from the real world
and the IBM customer (international consumer goods industry) who suffered
that 756 MB penalty for trying to us LPAR on his 6C4 with 4GB was really
p*ssed off because he could only use 3,25 GB per server in LPAR mode.
Actually he bought 4 servers with 4GB each and so lost 756 MB per server.
This amounts to about 3 GB of ram i.e. he lost almost one complete server's
memory. Hence he decided to use Full System Partition with two servers in
order to use the whole ram at least on those machines but at the same time
he was quite dissappointed of his IBM sales representative because that guy
did not sell him what he wanted. What can IBM learn from that? If a customer
wants a server with 4GB of RAM with an LPAR ready pSeries better make sure
that he not only gets 4GB but that he can *use* 4GB.

Regards,
Andreas

Hello,

> The other 512MB to 756MB bit is used for the Page Tables used to manage
> virtual memory (they are virtual address to physical address translation
> tables used by the CPU H/W).
> After an initial chunk, this is 1/64th of memory.
> This can be seen in LPAR machines because the Hypervisor is managing virtual
> memory to ensure the LPARs can never share the same physical memory (LPARs
> are isolated) abd is seen on one of the HMC machine proprty panels.
> Exactly, the same amount of memory is used when not in LPAR mode - i.e. SMP
> mode or just running on a normal RS/6000 or pSeries machine but you are
> never told how much memory is used by these same Page Tables.
> So yes the 512MB is an overhead you pay for LPAR but the rest is not.

there exists one page table per process, doesnt it?

So, this hypervisor page-table is needed additionally, isnt it?

I am not really sure - but somehow I feel hard to believe that such a
typical os-task like maintening pagetables has been moved completely out
of the os.

Regards,

Holger

"Holger van Koll" <holger@vankoll.de> schrieb im Newsbeitrag
news:1087301023.842821@ftpgate...
> Hello,
>
> > The other 512MB to 756MB bit is used for the Page Tables used to manage
> > virtual memory (they are virtual address to physical address translation
> > tables used by the CPU H/W).
> > After an initial chunk, this is 1/64th of memory.
> > This can be seen in LPAR machines because the Hypervisor is managing
virtual
> > memory to ensure the LPARs can never share the same physical memory
(LPARs
> > are isolated) abd is seen on one of the HMC machine proprty panels.
> > Exactly, the same amount of memory is used when not in LPAR mode - i.e.
SMP
> > mode or just running on a normal RS/6000 or pSeries machine but you are
> > never told how much memory is used by these same Page Tables.
> > So yes the 512MB is an overhead you pay for LPAR but the rest is not.
>
> there exists one page table per process, doesnt it?
AFAIK the number of this page table (outside the partition memory) is
related to partition memory size and not to number of processes. It is 1/64
of the partition's logical memory rounded up to the next power of two. E.g.
an LPAR of 2048 MB has a partition page table of 32MB but an LPAR of 2112 Mb
has one of 33MB and hence takes 64 MB (in one contiguous area of physical
RAM).
Because of that memory overhead of hypervisor, TCE and partition page tables
it might make sense on severs equipped with little RAM to define LPAR memory
size precisely to a number that allows optimal use of the partition page
table (just like you configure an LPAR size of 15756MB if you cannot use
16GB in one piece beginning on the 16GB boundary (I.e. when you've got 32 GB
memory in your regatta)).

>
> So, this hypervisor page-table is needed additionally, isnt it?
Yes, this partition page table is additional memory that is necessary for
the partition to operate. Only what is left after the reserved memory
regions are taken away can be allocated by the LPAR itself.

>
> I am not really sure - but somehow I feel hard to believe that such a
> typical os-task like maintening pagetables has been moved completely out
> of the os.
Has it been moved completely out? I do not think that is likely. AIX VMM
uses the hypervisor to translate from virtual-to-logical mapping of AIX to
virtual-to-physical mapping of the Regatta within this page table.
What I am not sure about is the comment Nigel made about 'the same amount of
memory is used when not in LPAR mode'. If AIX uses a page table (for the
memory within the LPAR and uses another (?) page table (for the pSeries RAM
that is assigned to the LPAR) I would expect this to use more memory than a
page table of a pSeries in Full System Partition only. Maybe someone else
can shed some light on this.

Regards,
Andreas