R.S. wrote:
John S. Giltner, Jr. wrote:
[...]
Mostly Linux on the MF was so that you could migrate what was on 10,
20, 200 Intel boxes that were sitting there 90% idle and put on a
single box and share CPU and Memeory resources.
IMHO this is marketing mantra from IBM.
1. Sharing memory resource is not true. Memory can be easily
reconfigured (taken from system A to system B), but, assuming 24x7
availability it is big advantage.
2. Usually all the servers have workload peak in the same time. They
usually form single information system.
Depends on what type of servers and your environment. Our file servers
do not hit their peak at the same time as our print server, or e-mail
servers. Our Windows DC's almost never see about 10-20% cpu utilization.
3. Even If sharing resources could make some savings in term of
megabutes or CPU cycles, it doesn't necessarily mean any savings in
costs. CPU cycles and megabytes in PCs are much cheaper. Maybe cycles
are hard to compare, but memory price is simple to compare - megabyte
has the same meaning.
It also depends. If you know the distributed workload and when it is
busy and when it is not, you might be able to put many boxes on a single
mainframe. Remember, it is still recommended today that you do not put
more that one function on a single Intel server OS. You don't put DNS
serving on the same box you do print and file severing, or web severing.
You don't mix an application server and a DB server. So you end up
with a lot of boxes that have excess capacity. Not marketing from IBM,
fact. We have only have 40 distributed server. We have this because we
can't put print serving on the same box as the Windows Domain Control
function and we can put file severing on the same box as e-mail. The
MOST any one of these boxes CPU utilization gets is about 50% and that
is during backups. We could easily get 20 of these boxes on 2 IFL's but
we can't convince anybody to do it. Instead we are now going to spend
$100K on blade center with 4-5 blades to try and consolidate 10-15 of
these servers on to 4-5 blades using VMWare ESX.
Where it does come in handy is when you need hundreds of "servers" but
only a few , 10-100, are ever active at one time. Like in a collage
environment. Imagine giving 100 students 5 servers each. At Marist
they do this, with Linux on the mainframe. When you sign up for
computer courses you get 5 virtual Linux images that are yours for the
next 5 years.
4. There is also not mentione advantage of Linux on z/machine: hardware
reliability and availability. It is hardly comparable, the only thing in
replace of RAS could be hot reserve or clustering.
True, from an hardware stand point z boxes are the most reliable in the
industry.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
