Hi Carl,

Thanks for your input. 

The main reason I can see for not using a task is that you will have times
where an incident ticket is created for a service that's only related to an
OLA. So from a reporting perspective you would need to pull both incidents
and tasks related to the OLA, and from a SLM maintenance perspective you
would need to configure two different modules.

Here is my take on it.
 
Typically Service Level Requirements (SLRs) make up an SLA. These could be
statements like "submitted incidents will be supported 8am-5pm M-F",
"critical incidents will be responded to in 15 minutes and resolved in 1
hour", "the service will have 99% availability during working hours", and
"the application will process transactions within 5 seconds".
 
These would then be used to create the SLAs and service targets in Remedy.
Whether they are request-based service targets, availability service
targets, performance-monitoring service targets, or some other type is
dependent on the requirements, but all of those may be a part of the
agreement(s) with the customer.
 
OLAs are used for the services that support the SLAs and are internal to the
IT provider. One example could be where the app/dev team (creating a
customer relationship management system) consumes a service provided by the
server/platform team (the server and software that the CRM is being placed
on), which consumes a service from the database team (providing the
database, maintenance, and backups).

The customer may be using a customer relationship management service, but
behind the scenes that's made up of the various supporting services which
each team agrees to certain levels of service. So if a ticket is created
because the CRM is not working for the customer, then the cause(s) of the
issue can be found (bug in CRM code, server issue, or database issue) and
the appropriate team can resolve the problem and be held accountable to
violations of their OLA. One or more teams may need to help resolve and each
would have their specified amount of time to resolve their part of the
issue.
 
My point being I could imagine where the app/dev team becomes the customer
on the incident because they have an agreement with the server team to
provision an environment within a certain time frame, or correct an issue in
a test or development environment that does not affect an external customer.
So they would create an incident, relate a child/support service, and only
the OLA would attach to the ticket. In another scenario you may have an end
user report that the CRM is not working, which would spawn an incident with
the CRM service related to it and an associated SLA is attached. Then after
initial troubleshooting it could be determined that the database is the
culprit, so you would need a way to related the underlying Database Service
(OLA) to the ticket, in addition to the SLA. So it makes sense that we
really need a way to relate additional technical services to the incident in
order to have their related OLAs attach. 

Here's an example ( source
<http://wiki.en.it-processmaps.com/images/pdf/service_structure.pdf>  )

<http://ars-action-request-system.1.n7.nabble.com/file/n122026/wiki.png> 



--
View this message in context: 
http://ars-action-request-system.1.n7.nabble.com/Let-s-discuss-best-practices-on-SLM-between-the-SLA-and-OLA-relationship-for-BMC-Remedy-Incident-Mant-tp121967p122026.html
Sent from the ARS (Action Request System) mailing list archive at Nabble.com.

_______________________________________________________________________________
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
"Where the Answers Are, and have been for 20 years"

Reply via email to