HP EVA3000" vs IBM DS4300 Turbo

We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Can anyone help out? Has anyone used either of these before? Pluses
and minuses? Comparative charts/information anywhere?

Also, is it a fact that HP is getting out of the storage business. Is
this true? Thanks much for any help.

Why not create 2 LUNs instead ?
Just curious...

Jesus wrote:
We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Can anyone help out? Has anyone used either of these before? Pluses
and minuses? Comparative charts/information anywhere?

Also, is it a fact that HP is getting out of the storage business. Is
this true? Thanks much for any help.

Why not create 2 LUNs instead ?
Just curious...

I don't want to create 2 LUNs. It is my understanding that it is not
posible to reassign discs between the two LUNs - referring to Vraid.
Three of the servers are formatted Vraid1 (database) and the other is
Vraid5 (file server). Taking this into account, wouldn't it be
reasonable to create two different raids in two separate LUNs due to
partitioning requirements? I am probably not explaining myself well.
Thanks again.

On 12 Feb 2005 04:46:42 -0800, (Jesus) wrote:

We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Can anyone help out? Has anyone used either of these before? Pluses
and minuses? Comparative charts/information anywhere?

Also, is it a fact that HP is getting out of the storage business. Is
this true? Thanks much for any help.

Biggest problem with the EVA line is performance. To get that cool
virtualization you are talking about the drives all have to be behind
the same controller (and it's failover partner). This means you are
limited to the IO and bandwidth of that one controller for the LUN's.

In some instances the virtualization is the top requirement, in other
cases it's performance. When it's the latter the EVA line loses every
time.

~F

"Jesus" skrev i meddelandet
om...
We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Not quite. You create raidsets (0,1,3,5,0+1) out of drives. Then you create
LUNs from these arrays. I suppose you could create one big raid5 set out of
the 14 drives and carve LUNs from it. There's no true virtualization like
the EVA does where each LUN is it's own raidset across all drives.


Can anyone help out? Has anyone used either of these before? Pluses
and minuses? Comparative charts/information anywhere?


IBM's response to the EVA is to bundle an DS4x00 with a couple of SVCs (SAN
Volume Controllers). Basically a clustered in-band virtualization engine
running on a linux OS on intel. Quite nice, not sure of the stability but
looks good on paper. From what I've heard, about 1000 installation
world-wide.

Also, is it a fact that HP is getting out of the storage business. Is
this true? Thanks much for any help.

Doubt it very much.

/charles

Nothing but FUD.


On Mon, 14 Feb 2005 21:37:20 GMT, Faeandar wrote:

On 12 Feb 2005 04:46:42 -0800, (Jesus) wrote:

We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Can anyone help out? Has anyone used either of these before? Pluses
and minuses? Comparative charts/information anywhere?

Also, is it a fact that HP is getting out of the storage business. Is
this true? Thanks much for any help.

Biggest problem with the EVA line is performance. To get that cool
virtualization you are talking about the drives all have to be behind
the same controller (and it's failover partner). This means you are
limited to the IO and bandwidth of that one controller for the LUN's.

In some instances the virtualization is the top requirement, in other
cases it's performance. When it's the latter the EVA line loses every
time.

~F

--- jls
The preceding message was personal opinion only.
I do not speak in any authorized capacity for anyone,
and certainly not my employer.
(get rid of the xxxz in my address to e-mail)

On Tue, 22 Feb 2005 18:18:44 GMT, jlsue
wrote:

Nothing but FUD.


Uh, ok. Want to clarify which part exactly? I've had HP come in and
give their pitch on this and the one thing that is not in question
whatsoever is the single controller issue. Hell, they even admit to
that being a performance issue.

Now, given that, how do you think this box can compete with arrays
that can allow data access through multiple controllers?

~F

On Mon, 14 Feb 2005 21:37:20 GMT, Faeandar wrote:

On 12 Feb 2005 04:46:42 -0800, (Jesus) wrote:

We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Can anyone help out? Has anyone used either of these before? Pluses
and minuses? Comparative charts/information anywhere?

Also, is it a fact that HP is getting out of the storage business. Is
this true? Thanks much for any help.

Biggest problem with the EVA line is performance. To get that cool
virtualization you are talking about the drives all have to be behind
the same controller (and it's failover partner). This means you are
limited to the IO and bandwidth of that one controller for the LUN's.

In some instances the virtualization is the top requirement, in other
cases it's performance. When it's the latter the EVA line loses every
time.

~F

--- jls
The preceding message was personal opinion only.
I do not speak in any authorized capacity for anyone,
and certainly not my employer.
(get rid of the xxxz in my address to e-mail)

On 12 Feb 2005 04:46:42 -0800, (Jesus) wrote:

We are in the process of setting up a SAN of 2TBs and we are trying to
decide between these two systems. EVA3000 permits virtualraid. IBM?

For example, with 14 discs, EVA permits me to make a lun, partition it
in different raid types to use it in the distinct servers. IBM?

Just FYI, this is not exactly correct in how the EVA configures your
drives. You create disk groups and assign drives to the disk group.
Then you create individual LUNs at whatever size you need, and present
these to the hosts. When you create these LUNs, you specify what
VRAID level you want. All LUNs are spread across all disks in the
group, alleviating the most common bottenecks at the spindle level.

Now, one drawback of the smaller, 14-drive configuration that you have
is that you really can't get perfect protection from *all* failures,
because you have them all on one, or possibly two shelves. VRAID5
will protect you against a single drive failure, but it won't help you
in some other failure circumstances - rare though they may be.

Most all other storage subsystems I've seen are more bus-based in
their configuration - i.e., they don't have true virtualization of the
drives; and they don't automatically level the I/O load spread across
spindles to avoid hot-spots.

That's not to say that other controllers can outperform the EVA in
some workloads. However, many of these must use massive amounts of
cache to get that performance, and cache memory is pretty expensive.

--- jls
The preceding message was personal opinion only.
I do not speak in any authorized capacity for anyone,
and certainly not my employer.

On Wed, 23 Feb 2005 07:50:35 GMT, Faeandar
wrote:

On Tue, 22 Feb 2005 18:18:44 GMT, jlsue
wrote:

Nothing but FUD.


Uh, ok. Want to clarify which part exactly? I've had HP come in and
give their pitch on this and the one thing that is not in question
whatsoever is the single controller issue. Hell, they even admit to
that being a performance issue.

BS. Unless you test a specific workload, you do not know what the
performance characteristics of that workload will be.

In fact, the actual controller is not often the bottleneck as much as
the disk spindles.

In practice, having lots of spindles to service an I/O in a LUN on the
EVA will alleviate more bottleneck problems that most workloads see -
in my experience.


Now, given that, how do you think this box can compete with arrays
that can allow data access through multiple controllers?


The assumption is that the controller is the bottleneck. Something
which is not necessarily true, and especially at the 2TB EVA3000 level
that the original poster is considering.

All that said, the new EVA series announcements will greatly improve
this performance.
--- jls
The preceding message was personal opinion only.
I do not speak in any authorized capacity for anyone,
and certainly not my employer.

On Wed, 16 Mar 2005 14:10:27 GMT, jlsue
wrote:

On Wed, 23 Feb 2005 07:50:35 GMT, Faeandar
wrote:

On Tue, 22 Feb 2005 18:18:44 GMT, jlsue
wrote:

Nothing but FUD.


Uh, ok. Want to clarify which part exactly? I've had HP come in and
give their pitch on this and the one thing that is not in question
whatsoever is the single controller issue. Hell, they even admit to
that being a performance issue.

BS. Unless you test a specific workload, you do not know what the
performance characteristics of that workload will be.

That was direct from the HP engineers so take it up with them. IO
patterns are only a consideration when they don't require more than an
aggregate of a single controller, usually around 80MB/sec.


In fact, the actual controller is not often the bottleneck as much as
the disk spindles.

You're saying that 64 drives would be a bottleneck and not the single
controller in front of them? Right.


In practice, having lots of spindles to service an I/O in a LUN on the
EVA will alleviate more bottleneck problems that most workloads see -
in my experience.

A single LUN means a single system mounting it, generally. The shared
aspect of SAN means more than one LUN and more than system would be
accessing data behind that controller.



Now, given that, how do you think this box can compete with arrays
that can allow data access through multiple controllers?


The assumption is that the controller is the bottleneck. Something
which is not necessarily true, and especially at the 2TB EVA3000 level
that the original poster is considering.

Again, not an assumption but a fact stated by HP. You should talk to
them more without your rose colored glasses on.
Ask the hard questions that you apparently don't want the answers to.

I'm not saying the EVA doesn't have a place, the virtualization
capabilities of that single controller are actually cool. But as I
said, if performance is your main concern then the EVA is not on the
short list. Not by a long shot.

~F