ADSM-L

Re: new TSM Client Pricing - a BP opinion (continued)

2003-05-18 07:23:42
Subject: Re: new TSM Client Pricing - a BP opinion (continued)
From: Zlatko Krastev/ACIT <acit AT ATTGLOBAL DOT NET>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Sun, 18 May 2003 14:18:28 +0300
Again from me. Thanks to those who think my post was not so bad. For
others I would try to answer/clarify a bit more:

Processor-based licensing was introduced not only TSM but for all Tivoli
products after April'02 IBM "launch". This is marketing decision and not a
technical one. The developers definitely had no time to change the code.
For example I remember the discussion should the licenses be purchased for
each node on single system. The GUID should help on this.
I expect in new v5.2 processor-usage tracking to come, and some part of
this discussion to become obsolete.
BTW: I also recall someone on this list pointing his suggestion to the
IBM/Tivoli salesrep to go round and count the processors himself.

--> I will agree that the new pricing model is less than ideal

and I will second on this. But a lot of criticism without a better
alternative does not bring something useful. Remember: any new scheme must
be satisfactory for major part of both current and prospective customers
of IBM. Otherwise its implementation would be a suicide.

--> I can have an old multi-processors computer less powerfull than a new
mono-processor computer but I will pay more for the less powerfull one.

That's the price for your savings in hardware and migration. For example
you (or your colleague) can upgrade an old 4x Pentium@166 MHz with single
P4@2,8 GHz or replace 8-processor Wintel with 4-proc HP 9000 / IBM pSeries
/ Sun Fire, and pay less for licenses. It's up to you.
OTOH let imagine IBM/Tivoli will make you happy and make pricing based on
sum of MHz (like Oracle did few years ago) or sum of SPECint/SPECfp
numbers or sum of degrees Celsius of the processors' cores. How at that
moment you will calculate your usage for hundreds or thousands of nodes???
Or should buy more licenses if someone in the lab has overclocked his PC?
Afterwards at the price of making you happy IBM can raise anger in
thousands or hundreds of thousands other customers. Oracle got a wave of
complaints, and as some analysts said, lost marketshare, when MHz-based
licensing was introduced. I think that the whole industry learned from
their experience.

--> Your "past" does not go very far back ...

That's true. I was walking under the table without ducking when WDSF had
been on the market. But the things change, market changes, demands and
products change. Ford T was sold for $800. Can you buy any new car now for
same price? Will you be happy having only a wheel, four tires, engine and
brakes? Go and buy Backup Exec.

--> IBM had a program ... which allowed Universities ...

IBM and other vendors still make discounts for educational institutions we
can dream for.

--> client license was a flat price, and it did not matter what the client was
... (this has really not changed much, until V5, and the GUID).

This is not changed in v5 either. Each "client" is counted once whatever
type/size/power it is.
If it is service-providing system the license was counted as "Tiered" TSM
server or "Tiered" Managed System for LAN/SAN *TSM* client. Now it is
licensed by # of processors.
Counting an application, DB, file or other kind of server system as TSM
"Client" instead of MgSysLAN is license violation and is better not to be
announced in public forum like this one.

--> Universities are not run like Fortune500 companies. The CIO cannot dictate
to the researchers how they spend

They have to! And if someone does not want to spend on backup and is not
in the overall budget, for me means he/she does not need it.

--> ... type of organisation that buys M$ licences based on the number of
people working for it, completely ignoring ...

<MARKETING>
I would answer this is a good candidate for Tivoli Inventory (now part of
IBM Tivoli Configuration Manager) and I would be glad to discuss some
business with you.
</MARKETING>

--> Hector Chan

Try to start new thread and use meaningful subject and your chances to get
an answer would be better.

Zlatko Krastev
IT Consultant

P.S. Time for another nap. This one became long again :-)
ZK






Zlatko Krastev/ACIT <acit AT ATTGLOBAL DOT NET>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
14.05.2003 11:47
Please respond to "ADSM: Dist Stor Manager"


        To:     ADSM-L AT VM.MARIST DOT EDU
        cc:
        Subject:        Re: new TSM Client Pricing - a BP opinion


The thread is getting longer and longer and the frustration increases. So
I will throw another attempt to explain things.

1. Formal disclaimer:
I am working for an IBM Business Partner but not for IBM itself. The
opinion presented here is personal, binds our company but does not bind
IBM and is made at best will derived from IBM *official documents*. Many
of you do not have business with Bulgaria so I will skip the marketing.

...



"Klein, Robert (NIH/CIT)" <kleinr AT EXCHANGE.NIH DOT GOV>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU>
13.05.2003 20:10
Please respond to "ADSM: Dist Stor Manager"


        To:     ADSM-L AT VM.MARIST DOT EDU
        cc:
        Subject:        new TSM Client Pricing


I was speaking with our TSM sales rep earlier today about ordering TSM
client software upgrades.  She said that as of this past January, a factor
in the pricing of a client software license is whether or not the client
is
a server or a desktop and, if a server, how many cpus it has.  Has anyone
else heard anything about this?

Thanks.

<Prev in Thread] Current Thread [Next in Thread>
  • Re: new TSM Client Pricing - a BP opinion (continued), Zlatko Krastev/ACIT <=