I am looking for information on how large-scale database
administration organizations set service levels and charge for their
services.  We use the following DBMSs: Oracle, DB2, UDB, SQL Server
and IMS-DB.

My questions are:
1) Do organizations typically set different service levels for each
DBMS and do they vary by environment (i.e. production, system test,
training, development)?
2) What sorts of factors are used in setting service levels (e.g.
service interruptions)?  What are typical targets for these factors?
3) What basis is used for charging users for databases?

---

Request for Question Clarification by jbf777-ga on 06 May 2003 11:16 PDT

Hi 

Can you clarify your question a little?  What kind of large-scale
database administration companies?  For what kind of product?

jbf777-ga
GA Researcher

---

Clarification of Question by handa78-ga on 06 May 2003 14:13 PDT

Our organization is in the financial services industry.  We develop
and support applications that use the databases mentioned in the
original questions.  Our databases occupy about 9TB of storage.

We have two major roles: 1) develop databases as required by
enterprise projects, and 2) support databases in that are in use.  Our
questions are really around this second role, supporting databases
that are in use.

We have been measuring our service levels mostly around the number and
length of service interruptions.  We would like to know what other
organizations are doing in setting service levels.

As for charging for services, we have been charging by the type of
database and the amount of storage it occupies.  Now there is more
demand for high-availability database servers, for which we will be
charging a premium.  We are wondering what approachs other companies
are taking to charging for services.  Note that we are charging
internal customers only -- we don't sell our services to other
companies.