I’ve been following Netuitive for over two years now. You can do a search for them on my blog and see all of the various activities and how they’ve evolved over the years. I was very skeptical of their early claims such as “BSM by Lunch” which I’m glad they’ve now backed away from to focus on their core competencies and value add to the overall BSM solution stack. I wish they would have stuck with the blog they started at BSM Digest, but I understand the challenges.
The power of getting accurate, trusted events free from false positives and false negatives is CRITICAL to the underpinning of any good BSM solution. If you’re putting garbage on the dashboards of your tools that operations, support and executives have to see, you’re NOT going to be successful with your BSM strategy. I’m also now very interested in Integrien and ProactiveNet (BMC) and look forward to digging in deeper into their solutions. Nearly every client I’ve seen and even when I ran the monitoring tools group at EarthLink we all have the same problems that these vendors are addressing. They’re the ONLY ones filling these gaps as best I can tell.
I’m looking for a really good discussion on Netuitive’s Active Behavior Profile (ABP). Which of your nine patents apply to this concept. Does every managed element type have a unique ABP or does every actual component have their own ABP? If I want to model/manage a Windows 2000 server different from a Windows 2003 server, how does this work? Is this where Templates come into play. What data streams are “mashed up” in the ABP? Templates?
What vendors do you play best with? Where are the key details of how/what you leverage from each of these vendor solutions? Please share some details. If a client only has the out of the box hardware and OS monitoring using their vendor’s solution with out of the box configurations, what can I expect to see in SI? What will I be missing? Do you recommend certain things be turned on to get health, workload and other outputs? Please discuss. Is one vendor’s CPU or Memory treated the same as another vendors?
When Trusted Alarms are sent outbound towards an event management solution as SNMP Traps, do they include group and function information? What can be mapped into varbinds? How is the trap constructed? Where does this happen?
I’d love to see some hard tangible ROI discussed on how these products are helping. I’m also very interested to know if the typical reactive based operations and support organizations are ready to get more proactive based on what these three vendors can provide. Can they mature from the comfortable, reactive “it’s broke” world and operate in a proactive, predictive “it’s a problem in this area, trust me” world?
Look forward to the discussion!
Comments on this entry are closed.
Thanks for your interest in Netuitive. Netuitive’s software leverages existing monitoring agents, collecting raw numeric data at the sub-system metric level for each key performance indicator (KPI), such as CPU and memory utilization, context switching, disk and I/O activity and hundreds of other application metrics. Netuitive learns the behavior patterns of each individual KPI for any given day of week, hour of day and minute of the hour. It also automatically learns the interdependencies between KPIs and the degrees of influence between them.
This type of unbiased contextual analysis — which is made possible using multivariate regression and other mathematical techniques — is what differentiates Netuitive from the other analytics players and enables unrivaled accuracy. Through this approach our software is able to produce automated composite views of system and service health, as well as forecast issues up to two hours before they occur. Again, the software accomplishes this through its own observation, without human bias. The solution requires no manually set rules, scripts, thresholds or correlation mapping.
Netuitive software automatically determines the statistically relevant metrics for a given environment using its templates. So in the example you cited, a Windows 2000 server may be a different model than a machine with Windows 2003. More significantly, the software will learn the unique behavior profiles for each OS and/or application using dozens of KPIs. In fact, it’s likely that the behavior of one Windows 2003 machine may look very different from another, depending on its applications running and the particular usage patterns of the system. This is irrespective of the brand hardware, which is not really applicable since each system is treated as unique. The same applies for virtualized environments.
Regarding Netuitive’s “Adaptive Behavior Profiles,” each KPI is analyzed in three distinct ways based on real-time, expected and forecasted performance. So there are actually three Adaptive Behavior Profiles for each KPI. For a given system or application the software analyzes as many as 30 KPIs simultaneously to generate a single Health Index for the given element or application. All of these system Health Indexes are then cross-correlated across monitoring silos to provide an end-to-end view of service health, which rolls-up to our dashboard.
Regarding ROI, Netuitive’s self-learning performance management software automates much of the manual administrative tasks related to systems and service management; dramatically reduces MTTR with automated root-cause isolation; and brings IT organizations from a reactive to a proactive state. There is a video of one of our Fortune 50 customers telling their own story at http://www.accelacast.com/rich_media/gartner_netuitive/ , along with the perspective of a Gartner analyst.
I hope this helps.
Daniel Heimlich
Netuitive, Inc.