[linux-lvm] One VG or many?

[linux-lvm] One VG or many?

[Poul Petersen]

	We have about 2TB to slice up and using a single VG seems to provide
the most flexibility for usage of the space. However, using one VG seems
reminiscient of having "all your eggs in one basket" in that if the VG
becomes corrupt, all of the LVs are at risk. So, are there any situations in
which using multiple VGs would be preferable to a single VG, or vice versa?
I'm curious to find out if other people use multiple VGs and, if so, what
requirements or concerns lead to the decision and what criterion are used to
assign data to a specific VG.

Thanks,

-poul

Re: [linux-lvm] One VG or many?

[Stephan Austermuehle]

On Tue, Jun 04, 2002 at 02:03:38PM -0700, Poul Petersen wrote:

> We have about 2TB to slice up and using a single VG seems to provide
> the most flexibility for usage of the space. However, using one VG

First, I would separate system (= containing the OS) and data area.
Second, if you do not intend to stripe data across LVs I would create
one VG for every logical business/application unit and assign disk
space on demand. That means that you leave most disks in a pool and
if you need more space in one VG you'll add the disk online with
vgextend. Why not start with one disks per VG?

If you plan to use the striping feature of LVM you should think of
what is sane looking at the two aspects speed and cost of extension
(e.g., striping across four disks requires you to add four disks at a
time if you need more space).

Stephan

RE: [linux-lvm] One VG or many?

[Poul Petersen]

> First, I would separate system (= containing the OS) and data area.
> Second, if you do not intend to stripe data across LVs I would create
> one VG for every logical business/application unit and assign disk
> space on demand. That means that you leave most disks in a pool and
> if you need more space in one VG you'll add the disk online with
> vgextend. Why not start with one disks per VG?
> 

	I see how adding disks from a pool would work very well. The problem
with our arrangement is that we would like to combine disks into RAID 5 sets
and then use these RAID sets as the PVs. The optimal size of each RAID set
(optimal in terms of disk usage, controller redundancy, channel redundancy,
and enclosure fault-tolerance) is about 215GB; so we would have 8 PVs
available to use in VGs. This size seems insufficiently granular to use as a
"expansion unit" to each VG. In fact, even if we used the individual disks
in the array, they seem too large at 73GB. This is why we were thinking
about partitioning the RAID sets so they would appear to the host as smaller
disks which could be added to VGs (The disks, RAID sets and even partitions
are all externally managed on a Zzyzx RocketStor 2000).

	Let me ask a slightly different question - what is the advantage to
creating a VG for every logical business/application unit? Or perhaps, why
would I not want to use a single VG? Is it a good idea just in terms of
failure in one VG not wiping out everything? It would seem that a reasonable
analogy would be that one VG is like one disk with partitions (LVs) for each
business unit, whereas multiple VGs would be more like multiple disks. Both
seem to have sufficient ideological separation?

Thanks for you help,


-poul

Re: [linux-lvm] One VG or many?

[Goetz Bock]

On Wed, Jun 05 '02 at 12:33, Poul Petersen wrote:
> 	Let me ask a slightly different question - what is the advantage to
> creating a VG for every logical business/application unit? Or perhaps, why
> would I not want to use a single VG? [ ... ] It would seem that a reasonable
> analogy would be that one VG is like one disk with partitions (LVs) for each
> business unit, whereas multiple VGs would be more like multiple disks. Both
> seem to have sufficient ideological separation?
while I can not decide this for you (I'm just bored waiting for my Mac
to stuff this W2k image it took me 7 hours to create. Why the ... must I
work on a Windows only project with only Macintosh hardware? <rant/>)
I'm running 3 big storage areas (different project ;-)) with 400GB,
300GB and 260GB (big for me at last). All are RAID5 and one PV/VG each. 
If I need to add additional storagge (as it will happen next week) I'll
just get another x00GB RAID5 and make it a new PV for the existing VG.
I than sliche it up using LVs ad needed.
As the RAID5 provides basic failiure savety the real savety comes from
.
.
.
Tape Backup.

So unless I (belie I might) hit a fix PV/VG/LV size area I'd just go for
one VG.
(Well actually all my systems have 2VGs each: one RAID1 for system/os
and another RAID5 for data. okok, some are small and have only one VG on
RAID1 or no RAID at all (my "real" workstation), or are a PowerBook G4
:-( )

Cu,
    Goetz.

Re: [linux-lvm] One VG or many?

[Goetz Bock]

On Wed, Jun 05 '02 at 22:15, Goetz Bock wrote:
> So unless I (belie I might) hit a fix PV/VG/LV size area I'd just go for
> one VG.
Too many typos in here:
So unless I (believe I might) this a fix PV/VG/LV size barea I'd just go
for one VG.

Cu,
    Goetz.

Re: [linux-lvm] One VG or many?

[Stephan Austermuehle]

On Wed, Jun 05, 2002 at 12:33:23PM -0700, Poul Petersen wrote:

> Let me ask a slightly different question - what is the advantage to
> creating a VG for every logical business/application unit? Or

When I setup a system I try to separate data that doesn't belong
together as far as possible to get maximum protection from software
and/or hardware faults. Multiple VGs increase chances to let other
applications in different VGs continue to work if a PV in one VG
fails. Of course if a system only has one application or if it doesn't
make sense to split up the disks into multiple VGs then create a
single VG only. So maybe having multiple VGs is useful on large
servers only.

Stephan