×
The internal search function is temporarily non-functional. The current search engine is no longer viable and we are researching alternatives.
As a stop gap measure, we are using Google's custom search engine service.
If you know of an easy to use, open source, search engine ... please contact support@midrange.com.
This is a recurring theme: "Why can't my 'high priced' server from IBM
support commodity component 'x'?" In the past 'x' has been various tape
drives, 4mm and Travan come to mind, as well as other bits like PATA and
SATA disk drives. Let me try to shed a little light on why IBM usually
chooses not to support such devices.
First. IBM Does things that have an actual business case. Oddly enough
the IBM corporation is in business to make money so when they decide
what to support on Our system they look at how many people would
actually make use of such a feature. Then they look at the cost of this
support and this contributes to a decision to support or not support a
feature. Remember too that 'support' in the case of IBM i is more than
just creating a driver for a device - it's long term support of
configurations and use of the devices. But often this isn't even the
most important issue.
Second. IBM looks at the device and how it would be useful for Us. How
would we use it. So lets take a look at how for example a USB drive
would be used. The common question I've seen, and represented here, are
as a Backup device. So which drive do you have? I think I have that one
too, it's a "Cheap one." Yes it works and it has a USB2 interface
running at 480Mbps. Zing! But wait, that's only 50% more than ONE SCSI
drive in the system and only 16% of the speed of the current SAS drive
interface. Beyond that, the drive inside your USB enclosure likely isn't
anywhere near that fast. So even if it is useful it's a small pipe that
wouldn't be useful beyond very small systems.
Third. IBM looks at the devices and how will it perform in the IBM i
environment. They look at things like, oh I don't know, let's pick
RELIABILITY. In the commodity PC market you can get a pretty good sized
USB 'box' for cheap money. Is that really where you want the backup of
your world class server? When it fails, who you gonna call? You're gonna
call IBM. And all they can say is 'Well we didn't create the device, we
really can't help you." You of course will argue "But you Support it!"
and the battle begins. By using your own methods to get data to these
low end devices you get to use them and IBM gets to say: "YOU
consciously chose that method of backup to a non-supported device" and
therefore is off the hook for support. That may sound like a cop-out but
when the entire system is designed on Industrial Strength, World Class,
and Rock solid, there just isn't any great reason to support devices
such as this that are 'fraught with peril' (my words.)
Another twist is that IBM has to consider how YOU would use it. For
example others and myself have suggested that carrying around a huge
stack of CDs full of PTFs is a pain. So we postulated that having all
those images on a USB 'stick' or drive would be awesome! IBM could say
"OK we'll support them for that use as it makes sense." But they know
you'll use them for whatever bright idea you have. So they must consider
very carefully support if it could be misused and thus create a support
(and therefore customer satisfaction) nightmare.
So my final words: "Think before you ask." :-)
- DrFranken
On 2/9/2010 1:05 AM, Jacob Anderson wrote:
...but I was hoping that a system
manufactured in 2007 could mount a USB hard drive and write data to it. My
cell phone can do that, so why is it that a $22,000 server can't???
That's a jab at IBM. I hope someone from IBM reads this news group. There is
no reason not to mount a USB hard drive.
-- jake
As an Amazon Associate we earn from qualifying purchases.