NAEEM’s Blog @ Consultaneer

I have been thinking about for sometime now, what’s the appropriate combination/flow of these listed products for customers, who want to achieve Business Service Management.

IBM Tivoli Monitoring (ITM), Tivoli Application Dependency Discovery Manager (TADDM) & Tivoli Business Service Manager (TBSM)
As we know, Business Service Management is not a product, but a concept, a combination of multiple products, which helps an Executive, CIO to know about the health of there services being provided to end user.
So let’s look at these building blocks towards a BSM case…
• First we lay the infrastructure components like network cabling, routers, switches etc.
• Then we put Servers, Printers etc.
• Once the systems are in place, Operating Systems like AIX, Linux, Solaris Windows etc. gets loaded
• Next comes the customer facing Applications, which sits on these Operating Systems
• Then the inter communications among these products (from user entering an order on web, all the way to confirmation of order – basically transactions occurring, which hits network, middleware, applications etc.)
As the business matures, things like historical collections, capacity planning, future projections, reporting comes into play.
Now as an IT shop goes live, we would like to know the Availability & Performance metrics, alerts and the Level of Service, our external / internal customers are getting (maintaining the Service Level Agreement (SLA) in place).
Data in form of events from Applications, Mainframes, Networks, Security, System & Transactions, can be gathered (using IBM Tivoli Monitoring (ITM) and its Agents) at an event management system like TEC, Omnibus for Availability & Performance. Normally operations team via there Event Consoles views these alerts or get paged and takes further actions to get in touch with an SME to drill drown or create automation to fix the problem. (It would be very interesting, if they are in pro-active mode, rather then reacting to situation, but that’s whole another discussion, on how to achieve that level of maturity, - In other words, reducing further MTTR, Mean Time To Repair)
Next comes the maintainability aspect of the IT Infrastructure
- Physical Infrastructure (Networks, Routers, Switches etc.)
- Hardware (Desktops, Servers, Storage, Mainframes etc.)
- Software (OS’s, patches, fixes, upgrades, New Applications)

Normally companies have some sort of CMDB (A configuration management database is a centralized and organized repository, containing information about whole IT Shop in theory), in forms of excel sheets, all the way to sophisticated relations databases, in which they track all these configuration changes. Basically in simple words, a giant database which is acts as an inventory for the IT infrastructure, hardware & software pieces.

Now as we know these systems are live and business running by the IT shop is dependent upon them. Without knowing the inter relationships, dependencies among these systems, applications, it becomes very risky to perform any maintenance. By bringing system A down, I can bring down Database A, on which multiple Web Applications are dependent running on other systems, thus causing downtime (which in business terms is, loss of $$$).
So in this case, TADDM (A tool to discover objects, show application mapping and  show data for change and configuration occurred to those discovered objects) can help, by scanning the whole IT shop and showing all the configuration and dependencies data in a graphical and table formats. Basically sketching a huge map of systems, devices showing arrows pointing to each other. User can then create the business services in TADDM to link systems, showing dependencies etc. These business services are basically a logical container of systems and applications running in the IT shop.
Once we business services information, we can upload them into TBSM. TBSM is a dashboard, where all data comes in to show the state of your service. So an as example, when a server goes down at Site A, I can view in TBSM dashboard, the impact of that server being down to the service being affected in the business.

So now back to my original thought, how should the flow work…

- Customers with Monitoring Software (for e.g. ITM), buys TBSM and later gets TADDM
- Customers with Monitoring Software (for e.g. ITM), buys TADDM and later gets TBSM
- Customers with Monitoring Software (for e.g. ITM) and wants to monitor there IT shop from a business level perspective.

For the last case, in my opinion it would make sense to create services tree, from a source which has all the relationship and detailed data about each and every component in an IT shop. In this case it will be CMDB, and since TADDM can get all that information and keep track of all the changes occurring in the environment periodically, it makes it a very viable candidate to feed services data into TBSM. (not to mention the Launch In Context (LIC), options between ITM, TADDM & TBSM)

So here’s a flow diagram I created, to show how the flow should look like.  

Let’s look at the major vendors providing Business Intelligence Software.

IBM, Oracle and SAP makes this list.

• Novermber 2007, IBM acquired “Cognos”for $US 5.65 billion, a compnay based in Ottawa, Canada.
• April 2007, Oracle purchased “Hyperion Solutions” for $US 3.30 billion, a company based in Santa Clara, CA, U.S.A
• October 2007, SAP bought “Business Objects” for $US 6.78 billion, a company based in Paris, France

The interesting point for me by these acquisitions is, how they can fit in the ESM (Enterprise Systems Management) market. As we in ESM tend to collect tera bytes of data. How can we bring intelligence out of that humongous amount of data and how well these applications will deal with TDW data. Also how the Cognos application will integrate with TBSM (IBM Tivoli Business Services Manager) and data collected via ITM (IBM Tivoli Monitoring).

BSM is a great area to look into and the timing is absolutely correct now, as this BSM idea has been around for sometime.
Being new to this, the biggest challenge I see from reading various articles and by doing the actual implementation of these products (ITM, TADDM, Omnibus, EIF, TBSM) is how to make get business sense out of all the data provided by these tools in a nice executive dashboard or in a very simpler, meaningful manner to end user.
As from my last 2 experience driving major part of ITM Best Practices, via Whitepapers, UA solutions and visits to many customers all around the globe, I think the need for “Best Practices” for BSM space is vital. A forum/team which can show users, using step by step scenarios and uses cases, how TBSM can fit in there environments.
The simpler we can present this idea, the more impact it will have.
Starting with a nice simpler scenarios of ITM with TBSM, involving TBSM.
Then we can go into Capacity Planning tools (Performance Analyzer), SLA, TADDM & CMDB in the picture, as we move forward.
So if I am a customer and wants to get this Solution (ITM/TBSM/Omnibus), I need a simple paper, explaining how the Events from ITM going to TBSM via Omnibus, can show my business picture. So this will require creating some meaningful “Services” (which is also a big area to look into, how to create an efficient service/rule).
So anyways, just my few cents as I am new to this and will be deep diving into BSM and driving Best Practices for BSM.

This blog will feature the areas of Enterprise Systems Management & Business Services Management as well as the Quality Assurance Aspect of Software Business.