On Thu, 2007-08-02 at 00:22 +0200, Lindy Mayfield wrote:
Does that mean that potentially that 1 mainframe has the equivalent of
at least 130 network cards? I can see how most of the hardware is
virtualized, but the networking I don't quite see, yet. How does that
part work?
As an option to
Does that mean that potentially that 1 mainframe has the equivalent of
at least 130 network cards? I can see how most of the hardware is
virtualized, but the networking I don't quite see, yet. How does that
part work?
It's a common networking technique called VLAN trunking; the network
not randomly pick these 3,900 servers.
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Lindy Mayfield
Sent: Wednesday, August 01, 2007 6:22 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: FW: IBM saves $250 million consolidating Linux servers on
to mainframes
-390@VM.MARIST.EDU
Subject: Re: IBM saves $250 million consolidating Linux servers on to
mainframes | NetworkWorld.com Community
On Wed, Aug 1, 2007 at 6:41 PM, in message
[EMAIL PROTECTED], Alan Cox
[EMAIL PROTECTED] wrote:
-snip-
Now does anyone know what percentage of the IBM saving
On Thu, Aug 2, 2007 at 9:22 AM, in message
[EMAIL PROTECTED], David Andrews [EMAIL PROTECTED]
wrote:
-snip-
As an option to communicating through an OSA, virtual machines can also
communicate via so-called hipersockets -- which are in essence OSA
emulators written in firmware. The TCP/IP
On Thursday, 08/02/2007 at 10:34 EDT, Britz, Anton - CO 7th
[EMAIL PROTECTED] wrote:
So you want to merge 4000 Servers on to ONE IBM box and IBM them selves
can not even keep their support system up..
This is how rumors get started. No one is merging 4000 servers on one
box. Don't get me
Hypersockets is best communication between LPARs. For communications
between virtual machines within the same z/VM LPAR, virtual machines
connected to the same VSwitch can talk directly to each other (still no
wires).
Rodriguez, Oscar wrote:
Does that mean that potentially that 1 mainframe
-Original Message-
From: Linux on 390 Port on behalf of Rich Smrcina
Sent: Thu 8/2/2007 11:14 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: FW: IBM saves $250 million consolidating Linux servers on to
mainframes | NetworkWorld.com Community
Hypersockets is best communication between
It is common to find a fair part of a server's network bandwidth used
for network backup and other service activities. If you can run the
backup server on the same z/VM system and connect them all to vswitch,
then all that network traffic can stay inside z/VM. If you used
dedicated OSA devices,
-Original Message-
From: Linux on 390 Port On Behalf Of Britz, Anton - CO 7th
Hi,
Yes, they should rather have publish the story that IBM's
IBMLINK is now down for how long ? A week...
So you want to merge 4000 Servers on to ONE IBM box ...
One and twenty-nine more.
and
According to the hardware manuals, a z9 can support up to 24 OSA cards;
with each card having 2 ports, for a total of 48 ports. Each port is
rated at 1 Gbps (I think that's correct), so I believe the aggregate
total bandwidth should be something on the order of 48 * 1 Gbps or a
total of 48 Gbps
On Thu, 2007-08-02 at 09:14 -0600, Mark Post wrote:
On Thu, Aug 2, 2007 at 9:22 AM, in message
[EMAIL PROTECTED], David Andrews [EMAIL PROTECTED]
wrote:
As an option to communicating through an OSA, virtual machines can also
communicate via so-called hipersockets -- which are in essence
According to the hardware manuals, a z9 can support up to 24 OSA
cards;
with each card having 2 ports, for a total of 48 ports. Each port is
rated at 1 Gbps (I think that's correct), so I believe the aggregate
total bandwidth should be something on the order of 48 * 1 Gbps or a
total of 48
http://www.networkworld.com/community/node/17998
--
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
On Wed, Aug 1, 2007 at 8:53 AM, in message
[EMAIL PROTECTED], McKown,
John [EMAIL PROTECTED] wrote:
http://www.networkworld.com/community/node/17998
It's too bad the article repeats the idea that's going around that Red Hat's
recent agreement with IBM is something new. Red Hat was later to
Post
Sent: Wednesday, August 01, 2007 10:59 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: FW: IBM saves $250 million consolidating Linux servers on
to mainframes | NetworkWorld.com Community
On Wed, Aug 1, 2007 at 8:53 AM, in message
[EMAIL PROTECTED],
McKown,
John [EMAIL PROTECTED] wrote:
http
-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Mark Post
Sent: Wednesday, August 01, 2007 10:59 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: FW: IBM saves $250 million consolidating Linux servers on
to mainframes | NetworkWorld.com Community
On Wed, Aug 1, 2007 at 8:53 AM, in message
, but the networking I don't quite see, yet. How does that
part work?
Lindy
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Kelman, Tom
Sent: 1. elokuuta 2007 20:49
To: LINUX-390@VM.MARIST.EDU
Subject: Re: FW: IBM saves $250 million consolidating Linux servers
Linux on 390 Port LINUX-390@VM.MARIST.EDU wrote on 08/01/2007 05:22:23
PM:
SNIP
Does that mean that potentially that 1 mainframe has the equivalent of
at least 130 network cards? I can see how most of the hardware is
virtualized, but the networking I don't quite see, yet. How does that
million consolidating Linux servers on
to mainframes | NetworkWorld.com Community
z/VM simulates the function of a hardware switch (called the Virtual
Switch). Each virtual machine (and even TCP/IP on VM itself) can use
the virtual switch to access the 'real' network. The traffic from the
machines
z/VM simulates the function of a hardware switch (called the Virtual
Switch). Each virtual machine (and even TCP/IP on VM itself) can use
the virtual switch to access the 'real' network. The traffic from the
machines on the virtual switch can be load balanced (as of z/VM 5.3)
over 3 OSA Express
for your cooperation.
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of Lindy
Mayfield
Sent: Wednesday, August 01, 2007 3:22 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: [LINUX-390] FW: IBM saves $250 million consolidating Linux
servers on to mainframes
:[EMAIL PROTECTED] Behalf Of
Lindy Mayfield
Sent: Wednesday, August 01, 2007 3:22 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: FW: IBM saves $250 million consolidating Linux servers on
to mainframes | NetworkWorld.com Community
Just curious, because I don't know how the hardware works, if 30
mainframes
Does that mean that potentially that 1 mainframe has the equivalent of
at least 130 network cards? I can see how most of the hardware is
virtualized, but the networking I don't quite see, yet. How does that
part work?
They are virtual - how many network cards do want - just write some more
Just curious, because I don't know how the hardware works, if
30 mainframes do the work of 3,900 servers, that means 1
mainframe does 130.
Does that mean that potentially that 1 mainframe has the
equivalent of at least 130 network cards? I can see how most of
the hardware is virtualized,
: IBM saves $250 million consolidating Linux servers on
to mainframes | NetworkWorld.com Community
z/VM simulates the function of a hardware switch (called the Virtual
Switch). Each virtual machine (and even TCP/IP on VM itself) can use
the virtual switch to access the 'real' network. The traffic
Just curious, because I don't know how the hardware works, if 30 mainframes do
the work of 3,900 servers, that means 1 mainframe does 130.
The big difference is that servers are running under 20% (in general).
The mainframe is running at 100%.
Does that mean that potentially that 1 mainframe
On Wed, Aug 1, 2007 at 6:22 PM, in message
[EMAIL PROTECTED], Lindy
Mayfield [EMAIL PROTECTED] wrote:
Just curious, because I don't know how the hardware works, if 30
mainframes do the work of 3,900 servers, that means 1 mainframe does
130.
Does that mean that potentially that 1 mainframe
On Wed, Aug 1, 2007 at 6:22 PM, in message
[EMAIL PROTECTED], Lindy
Mayfield [EMAIL PROTECTED] wrote:
Just curious, because I don't know how the hardware works, if 30
mainframes do the work of 3,900 servers, that means 1 mainframe does
130.
Does that mean that potentially that 1 mainframe
On Wed, Aug 1, 2007 at 6:41 PM, in message
[EMAIL PROTECTED], Alan Cox
[EMAIL PROTECTED] wrote:
-snip-
Now does anyone know what percentage of the IBM saving is computed by
calculating the saved advertising expenditure created by the story 8)
It's probably largely offset by the bonuses the
On Wednesday, 08/01/2007 at 06:43 EDT, Mark Post [EMAIL PROTECTED] wrote:
In addition to the VSWITCH that Rich talked about, each port on an OSA
card
provides between 15 and 48 network interfaces (depending on the model),
and you
can have multiple OSA cards. All but the OSA-Express2 10 GbE
Ahhh I was wondering how that worked since the DEFINE VSWITCH
command still only allowed three device numbers to be coded. Then I saw
the doc on the SET PORT GROUP command.
So 8 it is...
Alan Altmark wrote:
On a z9 with z/VM 5.3, a single VSWITCH can aggregate up to 8 dedicated
OSA ports
32 matches
Mail list logo