RE: iSeries TCO v NT server farm

>Yeah, but let's be clear here: "rolling upgrades" means a completely
>redundant set of servers...

No, no no, that's exactly not what it means. Let's say I have 2 servers
running my site. I can take one offline, leaving 1 server running my
site, upgrade that one and then put it back online. Then I can take the
other server offline and upgrade it and put it back online. If I have
only 1 server, then yes, I need a completely redundant server, but once
I have more than one I can just use part of the cluster for the upgrade.

>I'd love to see the offsite DR system that can handle
>20 Windows servers--or 120... and how much that costs.

We use one in Hawthorne, NY. We've got a dozen or so servers up there
and that's not a drop in the bucket for that data center. Costs a couple
grand a month -- peanuts.

>3. You can partition your machine and use a small partition as the
>"backup" box as in option 1.  

Um, sort of. Yes, you can do an OS upgrade on partition 1 while running
partition 2. But what about hardware upgrades, or upgrades to the
service processor?

>the vast majority of software upgrades for the iSeries can be 
>loaded while the machine is running, and only a subset of them 
>require so much as an IPL

No argument -- and the same is true of windows. I'm speaking of the
upgrades that do require downtime. How long does it take to upgrade the
OS? Between full save (twice?), OS load, ptf load, validation and full
save, that's usually a full weekend event -- often a long-weekend. That
really puts a dent in the five-nines concept. 

>Personally, I doubt there are a lot of truly 24/7 sites.

I'll agree with that!

>POINT: iSeries upgrades are less frequent, less intrusive and require
>less overall downtime than Windows upgrades.

On a single piece of hardware, perhaps -- and I'm not convinced about
that. However, that misses the point. Users no longer care if the
computer is up or down, they care if the application is up or down.
Multiple pieces of hardware will _always_ provide better uptime than a
single piece.

-Walden

PS. It's the long weekend. How many people are doing OS upgrades this
weekend?

------------
Walden H Leverich III
Tech Software
(516) 627-3800 x11
WaldenL@xxxxxxxxxxxxxxx
http://www.TechSoftInc.com

Quiquid latine dictum sit altum viditur.
(Whatever is said in Latin seems profound.)


-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Joe Pluta
Sent: Friday, September 02, 2005 3:15 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: iSeries TCO v NT server farm

Yeah, but let's be clear here: "rolling upgrades" means a completely
redundant set of servers.  You can't have, say, five servers and then
one that backs up the other five.  So the concept of redundancy on a
server farm is no different than redundancy on an iSeries... except that
with an iSeries you have many more options:

1. You can have a smaller backup box dedicated only to mission critical
functions that is much cheaper than your primary box, and is used during
outages.  My guess is that maintaining such a system in a Windows
environment is a bit more difficult.

2. You can have an offsite box that you lease from a DR firm for major
outages.  With creative planning, you can conceivably use this box for
planned outages.  I'd love to see the offsite DR system that can handle
20 Windows servers--or 120... and how much that costs.

3. You can partition your machine and use a small partition as the
"backup" box as in option 1.  This is less amenable to unplanned
outages, but works very well for planned maintenance and can be a
cheaper option (although with the price of a 500CPW machine down around
$10K, it's harder for me to justify the cost and complexity of
partitioning).

POINT: A single machine is easier to maintain, upgrade, manage and
failover than a cluster.  A cluster provides no more failover capability
than a single machine; you still need redundant machines, connections,
etc. in order to successfully provide HA.


Also, I don't know as much about Windows, Walden, but the vast majority
of software upgrades for the iSeries can be loaded while the machine is
running, and only a subset of them require so much as an IPL, which is
indeed a wee-hours planned outage thing.  Personally, I doubt there are
a lot of truly 24/7 sites.

Frankly, I see "system unavailable" messages on even supposedly 24/7
sites ("The electronic banking system is unavailable for planned
maintenance") with far more regularity than I have ever seen an iSeries
IPLed.

POINT: iSeries upgrades are less frequent, less intrusive and require
less overall downtime than Windows upgrades.


Joe


> From: Walden H. Leverich
> 
> >Most importantly, business is 24x7.  This application, once live, can
> >probably withstand short downtimes in the wee hours each day, but not
> >much more.
> 
> Rich,
> 
> Sure, the iSeries is stable, but it does go down. Sometimes (granted
> rarely) in an unplanned way, but other times in a planned way, be it
for
> hardware upgrades, OS upgrades, full backups, PTF installs, etc. And
> these planned downtimes are often more than "short downtimes in the
wee
> hours each day."
> 
> So, if you need to be up nearly 24x7, are you planning on buying two
> iSeries?
> 
> Part of the advantage of a cluster is so the application can survive
> "unplanned" outages, but the other advantage is the ability to do
> rolling upgrades where you take a part of the cluster offline and
> upgrade it, but the application continues to run on the other parts.
> 
> Just something to add into the mix.
> 
> -Walden