John,

In my opinion it sounds like you guys are over thinking this a bit and making 
it way too complicated.

Let's say you have a 4 hour resolution time SLA on "Email Service" related 
tickets:


-          A 4 Hour Resolution Time SLA would be setup - commitment to your 
customers

-          Now you need OLAs and UC that will ensure the SLA is met.

-          Let's say you give tier 1 one hour to troubleshoot (assigned to) 
before it goes to tier 2 - 1 hour OLA

-          Tier 2 has 1 hour - 1 hour OLA

-          Tier 3 has 1 hour - 1 hour OLA

-          Vendor support has 1 hour - 1 hour UC

This would give each group time to troubleshoot in order to meet the SLA.

Now there may be a bunch of reasons why email is not work, like "Network 
Services".  I would imagine that if Network Services is having issues then 
there are a lot more services impacted besides Email.  Network Services may 
have a different resolution time SLA.  I would think this needs to be treated 
complete separate from Email, even though it impacts email.  Hence the 
recommendation on Incident/Problem relationships.  In essence you would put the 
blinders on for each service.  Otherwise I think you're going to find out you 
have a very complicated mess that changes as the wind blows.  If I was in that 
position I would try really hard to talk the decision makers out of going to 
the complicated path.

Brian


From: Action Request System discussion list(ARSList) 
[mailto:arslist@ARSLIST.ORG] On Behalf Of John Johnson
Sent: Monday, June 20, 2016 9:55 AM
To: arslist@ARSLIST.ORG
Subject: Re: Let's discuss best practices on SLM between the SLA and OLA 
relationship for BMC Remedy

**
Thanks everyone. I agree with the concepts laid out by Carl and Brian, and 
those concepts are basically what I was thinking just written in a different 
way.

The problem I'm having has to do with how to implement these concepts in Remedy 
in a consistent manner. A lot of this has to do with SRM.

With SRM being the point of interaction between the customer and IT, this is 
ideally where the SLA/SLTs should be attached. For tickets that are created 
through SRM, it's fairly easy for us to relate a specific Service and SLT. This 
allows you to measure the full time from creation of the service request, the 
time until approval (if there is an approval), and the time to complete from 
approval. Then you could also attached the OLA SLTs to the fulfillment 
application(s).

The area I'm concerned about is when a ticket is created in the fulfillment 
application first and then the related service request is generated from there. 
There's no way to specify the related service, and service targets, that I'm 
aware of. So the process is not consistent.

There's also the concern that the related Service on the fulfillment ticket 
could be different. Are they picking the business service (parent service) and 
the technical service (child service)? For me, it seems like we should be able 
to relate more than one Service to a ticket. Something like one Business 
service and multiple technical services.





On Fri, Jun 17, 2016 at 6:26 PM Joe D'Souza 
<jdso...@shyle.net<mailto:jdso...@shyle.net>> wrote:
**
That is a very good explanation of the relationship between SLA's and its 
related child targets OLA's and UPC's.

>From an implementation perspective if you choose to use SLM, consider OLA's 
>and UPC'sto be subset of SLA's for every task that needs to be performed to 
>protect the main SLA. So if you will an OLA and UPC is like an SLA to every 
>task that needs to be performed by an internal or external group respectively.

Joe

________________________________
From: Action Request System discussion list(ARSList) 
[mailto:arslist@ARSLIST.ORG<mailto:arslist@ARSLIST.ORG>] On Behalf Of Carl 
Wilson
Sent: Friday, June 17, 2016 4:39 PM

To: arslist@ARSLIST.ORG<mailto:arslist@ARSLIST.ORG>
Subject: Re: Let's discuss best practices on SLM between the SLA and OLA 
relationship for BMC Remedy

Hi,
Normally the following applies:

SLA - Agreement with the Customer for a defined Target with penalties/rewards 
for late/on time delivery.
OLA - Agreement between delivery teams that are fulfilling the request e.g. 
Support Groups such as L1, L2, L3, etc. [this should be a product of the 
overall SLA i.e. the OLA's should not exceed the overall SLA to the customer, 
but should be a breakdown of the total SLA].
UPC - Agreement with a third party vendor i.e. External provider.

So, in your example:

1.  A customer reports a Service that is delivered to them is down "Email".  
The ticket is created that carries an overall "SLA" (agreement with Customer on 
defined Service Targets for up time e.g. 95%).  The moment the ticket is 
raised, an SLA is applied and starts measuring.
2.  A Support agent is assigned and works the request.  It may transition 
between internal support groups to resolve e.g. Level 1 (Support Desk), Level 2 
(SMA - Subject Matter Expert), Level 3 (Engineer, etc).  The OLA's form the 
basis of the SLA e.g. if a SLA has 8hr resolution time, each time may have a 
defined period of time to resolve i.e. Level 1 = 4hrs, Level 2 = 3hrs, Level 3 
= 1hr (Total = 8hrs = SLA).  You would therefore base the "OLA" targets based 
on the Support Groups. (not the overall "Status" of the request which would 
determine the SLA e.g. Start = Assigned, Stop = Completed).  Each OLA would 
start when the Assigned Group changes, therefore measuring the time each group 
spent working the request.
3.  If a third party vendor is required to be involved, then you may "suspend" 
(pend) the SLA/OLA until the third party vendor resolves the issue, where the 
SLA/OLA would stop and not continue until the vendor has completed the work.

In the above, the OLA's per team are a product of the SLA defined with the 
customer i.e. you could not have a total OLA (combination of all OLA's) that 
are longer than the overall SLA.  The OLA's are a breakdown of the total SLA.

The "Service" should define the SLA, the interaction between resolver teams 
will determine the OLA's that are defined between the resolver teams with 
respect to the SLA e.g. a breakdown of the SLA into periods that each resolver 
team has to perform their work with respect to the overall SLA (not exceeding 
the overall SLA).

----------------------------------------------

Kind Regards,

Carl Wilson


From: Action Request System discussion list(ARSList) 
[mailto:arslist@ARSLIST.ORG<mailto:arslist@ARSLIST.ORG>] On Behalf Of Brian 
Pancia
Sent: 17 June 2016 19:58
To: arslist@ARSLIST.ORG<mailto:arslist@ARSLIST.ORG>
Subject: Re: Let's discuss best practices on SLM between the SLA and OLA 
relationship for BMC Remedy

**
Looking at a pure ITIL approach.

Process Started

Incident - Email not working; Email SLA/OLA attached

Incident sent to next tier, where they determined it is a network issue.  At 
this step a Problem ticket would be opened, the Incident would be related, and 
any SLA/OLA for the Problem ticket would be generated.

More than likely the next step would be to issue a Change Request to fix the 
network issue, which in turn may have additional SLA/OLA attached.


Good Luck,

Brian


From: Action Request System discussion list(ARSList) 
[mailto:arslist@ARSLIST.ORG] On Behalf Of John Johnson
Sent: Friday, June 17, 2016 1:30 PM
To: arslist@ARSLIST.ORG<mailto:arslist@ARSLIST.ORG>
Subject: Let's discuss best practices on SLM between the SLA and OLA 
relationship for BMC Remedy

**

The more I work on trying to provide best practices to a customer on SLM, the 
more I feel like I'm just not there in being able to fully explain the concepts 
and correctly guide them on how to configure the SLAs and OLAs. There are two 
parts to my issue with SLM: the first part is with the concepts around SLM 
(services, SLAs, and OLAs) and the second part is with implementing them in BMC 
Remedy.  Please provide any comments you have on my thinking.


1.                              Concepts
1.                                                      Services
1.                                                                              
There are two main types of services: Internal/External Customer Facing 
Services (IT services) -- let's call them Parent Services and Supporting 
Services (IT Provider services)-- let's call them Child Services. The Child 
Services should support the Parent Services. Many Child Services may make up a 
Parent Service.
2.                                                      Service Level Agreements
1.                                                                              
These are agreements for the Parent Services
1.                                                                              
                        Example: Email
3.                                                      Operational Level 
Agreements
1.                                                                              
These are agreements for the Child Services
1.                                                                              
                        Example: Exchange, Network, Database
2.                              Configuration based on concepts above
1.                                                      Atrium Service Catalog
1.                                                                              
You would define your services
2.                                                      Service Level Agreements
1.                                                                              
You would create an SLA and relate a Parent Service
2.                                                                              
Regarding Incident Management, you would setup the SLA and then relate service 
targets to it. Typically the service target would only include the related 
company and priority in the "Terms and Conditions" field, and once related to 
an SLA, the service target would use the Service defined in the SLA as part of 
it's terms and conditions to get attached to an incident ticket.
3.                                                      Operation Level 
Agreements
1.                                                                              
You would create an OLA and relate a Child Service
1.                                                                              
                        Related to Incident Management, you would setup the OLA 
and then relate service targets to it. Typically the service target would only 
include the related company and priority in the "Terms and Conditions" field, 
and once related to an OLA, the service target would use the Service defined in 
the OLA as part of it's terms and conditions to get attached to an incident 
ticket.



So now my question is how would this work together in BMC Remedy? For example, 
a critical incident ticket is created because a user says that their email is 
not working and it's assigned to the Service Desk. The "Email" service would be 
related to the incident ticket using the "Service" field, which, in turn, would 
relate a specific service target (e.g. Email - Critical Priority Incident 
Resolution). Since "Email" is a Parent Service, the service target is related 
to an SLA. The "Email - Critical Priority Incident Resolution" service target 
will run from the time the ticket is created until it is resolved. The Service 
Desk does initial triage and then decides to escalate to a Tier 1 group. Tier 1 
determines that there is an issue with a network component. Ideally, this 
network component should be part of a child service, which should be related to 
an OLA.



So how should I continue? The ticket now needs an SLA and an OLA? Should a 
separate related incident ticket be created with the Child Service, so that the 
OLA would become attached? Is there some other way to have both the OLA and SLA 
attached to the original ticket? Should I design everything differently?



Another thought was that the SLA should only be attached to Service Requests in 
SRM, and then only OLAs would be attached to the fulfillment applications. If 
this seems like a better solution. My only question is, because I haven't 
tested it, can we relate SLAs to the generic SRDs that Remedy uses when a 
Incident User creates a ticket directly in Incident Management and the system 
creates the related Service Request?
_ARSlist: "Where the Answers Are" and have been for 20 years_

DISCLAIMER: The information contained in this e-mail and its attachments 
contain confidential information belonging to the sender, which is legally 
privileged. The information is intended only for the use of the recipient(s) 
named above. If you are not the intended recipient, you are notified that any 
disclosure, copying, distribution or action in reliance upon the contents of 
the information transmitted is strictly prohibited. If you have received this 
information in error, please delete it immediately.
_ARSlist: "Where the Answers Are" and have been for 20 years_
_ARSlist: "Where the Answers Are" and have been for 20 years_ _ARSlist: "Where 
the Answers Are" and have been for 20 years_
_ARSlist: "Where the Answers Are" and have been for 20 years_

DISCLAIMER: The information contained in this e-mail and its attachments 
contain confidential information belonging to the sender, which is legally 
privileged. The information is intended only for the use of the recipient(s) 
named above. If you are not the intended recipient, you are notified that any 
disclosure, copying, distribution or action in reliance upon the contents of 
the information transmitted is strictly prohibited. If you have received this 
information in error, please delete it immediately.

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

Reply via email to