One step at a time...

Running an enterprise IT department without an ITSM infrastructure is similar to running a finance department with spreadsheets. Ironically, IT departments support the 'computerization' of their company's business units that are running advanced solutions in the meantime, IT barely has any solution for it's own use. Now, when a customer has no ITSM platform and they ask us for reporting, I try to explain the steps needed to get there, that is infrastructure monitoring, service monitoring and then service reporting. Similar to setting up a financial system first and then expect to get your financial reports. If a customer has it's availability report then there is no beyond that as to the status and performance of the IT services. If I get a service report with 80% say availability, then I do know I have a problem but don't know why I have a problem and therefore I can't do much about it to improve it. Well, at that point I don't really need the report since the business units will be complaining about the IT service provided to them.
Therefore, the first step to implement an ITSM platform is the ITIL infrastructure optimization layer. Monitor every infrastructure component of IT and relate that component to the IT services provided. Focus on each IT service (on-line banking, pre-paid billing systems, ERP, etc. etc.) and then drill down to the infrastructure that supports the service rather than focus on servers and disks. Monitor every layer of the IT service, NSP, network, application, database, cluster, Operating System and hardware. Monitoring of only one layer, e.g. network with Network Node Manager, is only a part of the picture and will not bring significant value to IT and the business. Give the monitoring function to IT Operations since that is where it belongs, don't expect the adminstrators to monitor the infrastructure since that is not their job to begin with. This practice will increase availability since the time spend to find where the problem is (is it on the server, hardware or Operating System, is it the database, is it the network?) and reduce the finger pointing between the technical departments. You heard many times the phrase "it's not an application problem... it must be the database..." Well, eventually the problem is low swap on the server but you just wasted precious time of the application support administrator, the database administrator and the systems administrator to figure out something that a good ITSM implementation would have shown within 5 minutes. In the meantime, while the technical support personnel was searching and pointing to each other for 3 hours, your IT service was down!
Cars nowadays have advanced monitoring and diagnostic systems that let you know right away you have a problem, with the correct severity (red light for running out of oil, orange for check engine, cyan for running out of windshield fluid, etc. etc.) but your multimillion Euro IT infrastructure is running with no monitoring system.

IT Service Monitoring (end-to-end) is the next step. Try to monitor the IT service the same way the end-user 'sees' it. Emulate a user transaction and check that every few minutes in order to know right away you have a problem and record that data.
You are now ready to start working on Service Reporting.
Now, plug-in the bussiness requirements for the service (business hours, required availability, planned maintenance) and you got your report you need to do the job while Infrastructure monitoring will provide the data you need to take actions.

After you get to this point, then you can start looking into Business Technology Optimization, where you start monitoring flows of data, queues etc. etc. and translate that into business terms. In Infrastructure management, I have an alarm that the CRAPPQ queue is high and I let my application support people know about it to figure out where the problem is. When I get to business service management I translate this to 'credit application queue' and I link this to the rest of the business process supported by IT in order to receive and approve credit applications by customers.

Business Technology Optimization. Not everyone ready for it...

HP Software's direction is BTO (a term evolved from 'bussiness services management') that became a hot topic these days with the acquisition of Mercury. Starting from Infrastructure management and then off to service management the next logical step is business service management but I firmly believe that unless you have graduated the infrastructure and service management school, then you are not ready for business service management. A number of our customers are monitoring the IT infrastructure and services from the business point of view (meaning linking the infrastructure to the service and then managing the end service with monitoring, reporting and support). These kind of customers get into the business processs by monitoring the parts of the process. E.g. for banking payment systems there is a flow of the business process to accept for example utility payments either through on-line banking, ATM and other payment machines, process the payment and inform core banking, and other external systems and/or agencies of the payment status. Mercury has great tools to help us map application interaction and monitor the flow between them, which will help with the manual process we currently use to get people in a room and try to map the application flows on a whiteboard. How fast then can an organization move to BTO? That varies significantly, depending on a number of factors but unless these projects (infrastructure, service management) are backed by the top IT management then the expect a very slow pace graduating the ITIL arena with little projects that will win concensus and then move on to BTO.
George

ITIL Blog by HP's Jeroen Bronkhorst

This is my first posting to the blog, trying to get things started. I found an ITIL blog at HP by Jeroen Bronkhorst in conjuction with other HP folks, including Roc Paez. I remember meeting Roc at HP World in San Fransisco around 8 years ago, he was a bright young fella who did an excellent presentation on HP-UX.
http://h20325.www2.hp.com/blogs/itil

Rest of HP's blog can be found at:
http://www.hp.com/hpinfo/blogs/index.html

Happy postings,
George Petrides

Parallon Systems Blog

This is Parallon's blog for posting and sharing ideas with the rest of the world. This is a technology driven blog and we'll discuss IT infrastructure management, Service Management, ITIL and BTO with a business prospective.