Ok...so why do I think Operations Manager 2007 is probably the best product on the market at the moment for systems and service management? Well firstly before I begin I just want to say that I hope I don't come across as being too biased towards this product as there are many other systems management tools which each have good features which can be very beneficial. In order not to write an essay in one blog post I will split the reasons down into sectional blog entries - firstly the history.
To add justification it's important to know the background. A couple of years ago what was probably the most important aspect to many people wanting a systems management tool was the comprehensive monitoring of each part of their infrastructure e.g. how many performance counters could be monitored, how many rules for monitoring Exchange etc. So what was the problem with this? Well there were a number of issues with this:
1) Monitoring overload. Too many rules which were not pertinent to the environment being monitored which resulted in what we call a "sea of red" alerts which ultimately did not give any real value because actually everything seemed to be working ok...no one had phoned the support desk about any issues.
2) All monitored endpoints are equal. Well as we know this is not true - does a file and print server have the same importance as a Domain Controller? When a monitoring tool does not have any understanding of this we call this "hard monitoring" - any server goes down then a red alert is generated.
These are just two issues but there are many more and the result...well "if it don't work then flick the off switch" and generally this is what happened. So how does Operations Manager 2007 improve this and more specifically address the two issues above well:
1) Operations Manager 2007 Management Packs do have a lot of rules BUT they are very easy to 'tune' to the environment. There is no messing around with alerting monitoring scripts as all parameters are presented in the interface. Plus some threshold rules make use of Self Tuning Thresholds (STT) so baselines can be taken of your environment during a business week.
2) Operations Manager moves towards "soft" monitoring and more importantly service monitoring rather than system monitoring. All servers/systems work together ultimately to provide a service to an end user hence why service monitoring is important. No longer is a monitored endpoint just interpreted as a standalone component of an IT infrastructure but is part of a relationship to provide a service and Operations Manager 2007 is aware and monitors in this way through its entity health modelling and Distributed Application/Service Designer (DAD).
Of course there are many more historical problems in which Operations Manager 2007 can solve but his post is far too long already...so that's enough for now.