In the first major product refresh since closing the acquisition of IT service analytics firm ProactiveNet last summer, BMC’s new release, version 7.5, adds new autodiscovery and automated synchronization functions.

To recap, ProactiveNet adopts a self-learning approach to IT infrastructure performance, tracking performance patterns to detect potential problems before they erupt. By, in effect, “teaching itself” about usage patterns and changes to infrastructure and utilization it projects out into the future. BMC bought the company in spite of previously having a resale agreement with rival Netuitive, a newer, smaller 5-year old firm whose 2006 sales grew at a 400% rate. By comparison, ProactiveNet hit 56% year over year growth in 2006, but over a larger revenue base.

The highlight is a new data collection engine that, for now, is restricted to supporting the legacy Microsoft Operations Manager (MOM), rather than BMC’s own Performance Manager offerings. (Microsoft has since replaced MOM with Systems Center Operations Manager, a.k.a, SCOM). That’s because this was the enhancement that was in progress when BMC swooped in the buy the company last summer. In place of hard coded interfaces to MOM, it includes new features that can auto-detect and synchronize when new instances of MOM go live.

Similarly, it also adds some refinements for data filtering so you can make some changes to what data you want ProactiveNet to monitor without taking the system offline. And BMC has also added a data collection point for its own TMArt product, an offering that it OEMs from Borland’s SilkTest performance testing product line, which checks transaction response time.

Previously, ProactiveNet had a variety of data acquisition adapters to IBM Tivoli and BMC’s performance management tools, along with what used to be called CA Unicenter and HP OpenView that trapped either a database or some published API. But the interfaces or data acquisition engines were hardly standard.

Although BMC is not disclosing a detailed roadmap for ProactiveNet at this point, it goes without saying that the next enhancement for the product will be to add native support for BMC’s own Performance Management and Service Impact Management offerings. And down the pike, integration with BMC’s Atrium CMDB will be on the table.

Of course, all these enhancements simply skim the surface of what’s possible when you add predictive analytics to what has traditionally been a more backward-facing process, where you toil over data that trends past history or at best gives you a real-time snapshot of the present. A decade ago, BMC took the steps towards predictive analytics with its acquisition of BGS, whose technology is now the BMC Performance Assurance product. In that case, the predictives were employed strictly toward capacity planning, while with ProactiveNet, it is aimed at the different, although related task of maintaining service levels.

But this all begs the question that what good are predictive analytics if they are only used as a standalone solution? Ultimately, it would make sense to have a predictive analysis trigger in the background whenever any change is made to a system. And, ideally, such predictives should be translatable to higher level views, so that a business process such as order fulfillment could be forward tracked to see if a new promotion becomes so successful that it kills service levels. Or likewise, if your organization exposes a web service and offers a service level commitment, as to whether current usage patterns are likely to lead to an SLA compliance issue downstream. And all this ultimately impacts capacity.

Obviously, there are so many potential uses for predictive analysis that at some point, it no longer makes sense to offer it as a niche application that monitors certain consoles but not others. On the other hand, it doesn't always make sense to run such analyses because they can add significant overhead. The ultimate solution is having predictives available as a service that can be triggered based on policies.

---