Yes, I have heard something similar, John. The story I heard, and this
was about 3 or 4 years ago, is that IBM built a system that consisted of
a common backplane and disk drives, that would accept different types of
"servers on a card". If you wanted a zSeries box, you added zSeries on a
card to this system, if you wanted an iSeries or xSeries box, you simply
added those cards. The way cool part was that the different types of
cards could be mixed and matched in one system, with the backplane doing
the work so that the different cards could share the disk storage, and
communication amongst themselves at very high speeds (sounds like
Hipersockets between LPARs doesn't it?). You could have one system,
presented in a very tightly integrated manner, that could support both
zSeries and xSeries, say, workloads at the same time. I believe that
this type of approach is somewhat similar to the one now being promoted
by PSI, as well.
McKown, John wrote:
-----Original Message-----
From: The IBM z/VM Operating System
[mailto:[EMAIL PROTECTED] On Behalf Of Dave Jones
Sent: Wednesday, September 12, 2007 12:56 PM
To: IBMVM@LISTSERV.UARK.EDU
Subject: Re: z/VM and Linux in the news....
Bill, send me airfare and I'll be there! :-)
This does look like a really good meeting.....I am most interested in
hearing details of exactly how they are integrating the Cell
engine with
the zSeries mainframe (and z/VM, too). I suspect that, at the moment,
the integration takes the form of Linux guests under z/VM
sharing DASD,
via NFS or something similar, with the Cell-based blade servers (also
running Linux, btw). How the computational tasks are shared
amongst the
processors I haven't a clue....
DJ
MPI? Or one of the other "clustering" methodologies, I would guess.
Likely communicating via Gb Ethernet. The z/Linux instances might even
be communicating via NCSS shared memory. Or using some other z/VM
interface for inter-guest communications (might just be TCPIP on
hipersockets).
I still think that one technology that would be fantastic is one that my
boss insists IBM has. He says that IBM has a System z system on a card
that can be put in a blade server. This same server can also accommodate
pSeries and xSeries cards so that you can communicate between them on
the server's memory bus (like a hipersocket does between LPARs).
I do vaguely remember something like this for iSeries. Some sort of fast
interface to an xServer. The xServer, running Windows, could access the
iSeries' disk drives and talk to i5/OS on this "backplane" or something.
--
John McKown
Senior Systems Programmer
HealthMarkets
Keeping the Promise of Affordable Coverage
Administrative Services Group
Information Technology
The information contained in this e-mail message may be privileged
and/or confidential. It is for intended addressee(s) only. If you are
not the intended recipient, you are hereby notified that any disclosure,
reproduction, distribution or other use of this communication is
strictly prohibited and could, in certain circumstances, be a criminal
offense. If you have received this e-mail in error, please notify the
sender by reply and delete this message without copying or disclosing
it.
--
DJ
V/Soft
