Frequently Asked Questions

About Netuitive Self-learning Performance Management Software

What is the Netuitive Behavior Analysis Engine?

Built on nine patented technologies and techniques resulting from over 20 years of academic and commercial research and development, the Behavior Analysis Engine:

  1. Continuously analyzes and correlates thousands of simultaneous performance variables from dozens of enterprise sub-systems.
  2. Builds an Adaptive Behavior Profile for each data stream, reflecting the unique and changing rhythms of each performance metric — by the hour, month and season.
  3. Monitors the contextual relationships between each performance metric and applies real-time analysis and correlation to identify statistically significant anomalies.
  4. Delivers an actionable Trusted Alarm whenever it detects or forecasts changes in service levels.

back to top

How does the Behavior Analysis Engine learn?

Netuitive's patented Behavior Analysis Engine uses multivariate correlation and regression analysis algorithms, coupled with self-learning and adaptive capabilities, and a multitude of proprietary and patented statistical heuristics to process data in real-time. This ability is complemented by its rapid learning capability to adapt to sudden changes.

Netuitive's multivariate correlation analysis algorithm automatically learns, identifies and understands how two or more variables, or KPIs, co-vary in the natural environment. Correlation analysis techniques enable the engine to self-discover and study the relationships between variables. It understands how one KPI relates to the variability of another.

In addition, Netuitive uses multivariate regression analysis techniques to describe the numerical relationships between KPIs. Whereas correlation analysis intuitively surmises the relationship strength between two KPIs, regression analysis provides an algebraic equation describing the nature of the relationship. Furthermore, regression analysis provides the variance measures, which enables prediction and contextual analysis with imputed values.

The correlation and regression analysis algorithms are self-learning, meaning there is no initial bias when the system is configured. As a result, correlation coefficients and regression weights are based on the actual system behavior, not a pre-defined model.

Finally, the Behavior Analysis Engine is adaptive. In other words, as the engine is continually fed new data it automatically adjusts the statistical model (correlation coefficients and regression weights). Its models and dynamic thresholds are continually adapting.

back to top

What's unique about Netuitive's Behavior Analysis Engine?

In a nutshell: its speed, low processing and storage requirements and adaptability.

  • Speed — It takes only milliseconds to analyze thousands of simultaneous inputs.
  • Low processing requirements — A highly efficient approach enables Netuitive to perform, on a standard off-the-shelf server, massive calculations that would otherwise require a mainframe computer and databases.
  • Adaptive — The technology self-learns an environment's normal behavior, readily responds to changing IT requirements and automatically generates Dynamic Thresholds, and performs all of its computations with no manual configuration required.
  • Minimal storage — Netuitive uses a proprietary scoring method that checks data as fast as it is received and doesn't rely on large historical databases. One derived benefit of this patented capability is that Netuitive does not reprocess previously obtained data as new data is made available.

back to top

How does Netuitive determine which KPIs to monitor?

Netuitive uses both expert knowledge and statistical techniques to determine which KPIs are the best candidates to characterize statistical behavior of servers and other managed elements. Netuitive packages its best monitoring practices into templates. These templates represent the statistically proven and most relevant Key Performance Indicators for a given environment (e.g. UNIX, Oracle, Exchange, or Weblogic). Each template's pre-defined KPIs are then used by Netuitive SI and Netuitive Service Analyzer products to collect real-time performance data for generating health indicators and Trusted Alarms. It's important to note that a monitored element may require more than one template (e.g. Windows OS and Exchange Templates) to learn the intricate behavior of the environment. Customized templates can also be built using Netuitive Integration Studio.

back to top

How long does it take Netuitive software to generate behavior profiles and produce service health indicators?

It only takes two time intervals (of 1, 5, 10, 15, 30 or 60 minutes each, chosen for the deployment) for a baseline to start being defined. It takes a couple of more intervals for health information to be calculated. After just one day, you have a daily baseline that can be extrapolated to the remaining days of the week. After a week, your initial weekly baseline is completed.

Meaningful analysis results (health index, alarm index, workload indices, baseline, correlation coefficients, Trusted Alarms) often appear within the first couple of hours following installation, and the quality of the analysis continues to improve with time as the product learns. A waiting period of two weeks is generally recommended to obtain the most dependable analysis results. Finally, Netuitive performance analysis can be performed even at sub-minute frequency.

back to top

If my systems are performing poorly, won't Netuitive model the bad behavior?

In theory, yes. However, Netuitive's Behavior Analysis Engine has a built-in fast learning algorithm that detects abnormal change and can accelerate the pace at which environmental changes are learned. Bottom line: if detected problems are not addressed in a timely way, it can unlearn as fast as it learned a bad behavior.

Furthermore, Netuitive has complemented its purely mathematical approach by a set of user-defined policies and domain-specific heuristics and rules that prevent leaving a bad situation unalarmed. For instance, you can define a policy threshold forcing a deviation if the CPU Utilization exceeds a certain value (e.g. 95%), even if there is no statistical ground for that (e.g. low variability of CPU Utilization near the 100% ceiling). Similarly, you can define a policy filter that cancels any statistical deviation that happens in a considered-safe zone (e.g. below 30% for CPU Utilization), thus reducing the alarming noise.

back to top

How does Netuitive SI handle known performance peaks that are more periodic (e.g. end of month, holiday season, etc.)?

Netuitive is introducing a patented feature that enables the modeling of recurring or special events such as bi-weekly payroll, end of quarter activity, holidays or back-to-school sales when analyzing behavior patterns of IT systems. This functionality not only enables accurate analysis of a special event, but assures that the ongoing model is not polluted by known anomalies.

back to top

What is the difference between Netuitive SI and Netuitive Service Analyzer?

Netuitive SI analyzes the performance of infrastructure components (e.g. servers, clusters, farms, QoS transactions). Netuitive Service Analyzer analyzes the performance of IT services (e.g. on-line banking application, electronic retail), that is, inter-component analysis and relationship with end-user quality of service. Both Netuitive SI and Service Analyzer share the same underlying architecture and patented technologies, but they process elements at different levels of abstraction. Service Analyzer performs inter-component analysis and requires Netuitive SI for the data collection. Netuitive SI performs the intra-component behavior analysis.

back to top

What are the requirements for deploying Netuitive SI?

Netuitive SI has minimum installation requirements. It runs on several different platforms such as Windows 2000/2003, Linux 2.6, and Solaris 10. It requires three external components only: A relational database management system to persist its database (SQL Server, Oracle, Sybase, or MySQL), a browser (Internet Explorer or Mozilla Firefox), and one or more third-party monitoring solutions (e.g. BMC Performance Manager & PATROL, NetIQ AppManager, Microsoft MOM, HP, CA Unicenter) to supply performance measurements at regular intervals.

back to top

What are the requirements for deploying Netuitive Service Analyzer?

Netuitive Service Analyzer has minimum installation requirements. It runs on several different platforms such as Windows 2000/2003, Linux 2.6, and Solaris 10. It requires three external components only: A relational database management system to persist its database (SQL Server, Oracle, Sybase, or MySQL), a browser (Internet Explorer or Mozilla Firefox), and one or more instances of Netuitive SI to supply component analysis data. Netuitive Service Analyzer only integrates with Netuitive SI (or its OEM equivalents by BMC and NetIQ). When modeling a service in Netuitive Service Analyzer, it is important that Netuitive SI does behavior analysis of all infrastructure components that make up that service. In particular, the existence of QoS measurements for a service is key for a quality service model that offers a high level of prediction.

back to top

Where do Netuitive SI and Service Analyzer fit in my architecture?

Both Netuitive SI and Netuitive Service Analyzer are non-intrusive and have a very low impact on your existing monitoring system architecture. Netuitive SI does not deploy any new agents in your environment; instead it leverages performance data already collected by your existing third-party monitoring solution. The integration between Netuitive SI and your existing monitoring solutions is extremely lightweight and will not consume extensive network bandwidth, nor affect the workload of your monitoring solution (a limited number of performance counters is collected at regular intervals, e.g. every 5 or 15 minutes). Netuitive Service Analyzer does not touch your architecture at all. Its sole integration is with its companion product, Netuitive SI. If you already have a CMDB, Netuitive Service Analyzer may use its stored service definitions (via XML transfer), but this is not a requirement. Finally, both Netuitive SI and Netuitive Service Analyzer will seamlessly integrate with your event consoles and problem management solution (e.g. trouble ticketing system), by simply exporting its Trusted Alarms to them using a variety of standard (e.g. SNMP, SMTP) and proprietary protocols.

back to top

How does Netuitive Service Analyzer learn the "relationship" between service components without requiring user-defined dependency maps?

Netuitive SI provides Netuitive Service Analyzer with quality analysis data for the infrastructure components that make up a service. Each component is analyzed by Netuitive SI from three different perspectives: Availability, Performance, and Workload. This data is then processed by Netuitive Service Analyzer to self-learn and continuously adapt the correlation coefficients between service metrics. These coefficients make up the service dependencies and are used to determine service health and isolate root-cause in case of service outage.

back to top

Does Netuitive Service Analyzer require a CMDB?

No, Service Analyzer does not require a CMDB. A service is simply defined as a flat enumeration of the infrastructure components (servers, farms, clusters, QoS transactions) that make up that service. No topological relationships are necessary to define a service. Netuitive Service Analyzer performs correlation discovery that automatically self-learns and continuously adapts the dependencies between the service components. If you already have a CMDB, the existing service definitions could be imported into Netuitive Service Analyzer. This replaces the need to manually select the components in the Netuitive Service Analyzer user interface.

back to top

How does Netuitive's analytics approach differ from other IT analytics technologies?

Other behavior analysis tools use simple single-variable moving average calculations to determine dynamic thresholds for each individual metric, irrespective of the others. They typically rely on manual event scripting to assign correlation coefficients between variables, which are inherently inaccurate since they're based on guesswork. Netuitive software auto-configures and has a self-learning and adaptive bahavior analysis engine. It requires no manual scripting whatsoever. Netuitive's technology is based on multivariate "data" correlation and regression analysis as opposed to "event" correlation like most other techniques. Unlike other correlation products which rely on rules-based "events," Netuitive processes raw subsystem monitoring data to assure that alerts are based on actual behavior. Without Netuitive, an event correlation console is as good as the already-polluted events it is fed and relies on a human operator to define and script inference rules.

Furthermore, Netuitive products can complement existing rule-based inference engines. Too often, the rule-based systems are fed event streams based on static thresholds or single-variable moving average thresholds containing false-positives. As a result, by themselves they are not reliable (garbage-in-garbage-out). To the contrary, when Netuitive products are placed in front of an event correlation system, they will feed it a complete set of high-quality real-time and predictive events (Trusted Alarms), with no false-positives and no missed events, thus increasing the accuracy of rules-based systems.

back to top

How does Netuitive's approach to Performance Management and BSM differ from conventional solutions?

Traditional solutions rely on a labor-intensive and error-prone definition of service dependencies. This information is extracted from various subject-matter experts and systems and can be biased and incomplete. Instead, Netuitive Service Analyzer automatically self-learns and continuously adapts the service dependencies by exploiting the results of its patented correlation engine and the underlying infrastructure component analysis data (e.g. cluster health) generated by Netuitive SI. This approach is very fast, statistically accurate, and unbiased. The result is dramatically faster deployment, much lower administrative costs, and superior accuracy.

Netuitive ROI

back to top

Products