Application Monitoring

How do you decide what makes a great product release?  For me, there are two main drivers that bubble new features to the top of every release of RTView Enterprise Monitor:

1. Provide access to more actionable data
2. Facilitate ease of use

While it is an obvious step to include crucial missing metrics, the trick is to make that information something more than noise.  It has to be actionable.  Users need to be notified when conditions related to a component of the architecture are pertinent to application health and be able to easily perform deep analysis.  That analysis may be to determine how well the component is configured to handle load, how the component affects other elements, and ultimately how it will affect application health.

Ease of use is paramount when you consider that end-to-end monitoring systems handle massive amounts of information and therefore can be as complex as any application that they are monitoring. As organizations expand their portfolio of complex distributed applications, it can become a costly exercise to create a support team that can be effective at ensuring up time.  Often, you’ll need support staff members that are as highly trained in application architecture as the developers who created the applications.  These people are hard to find and expensive to employ. One of the most cost-effective ways to tackle this problem is to provide tools that are easy to use by less skilled support staff – at least for front line support.

 Let’s take a closer look at some of the new features of RTView Enterprise Monitor 1.5.

Solution Package for TIBCO ActiveMatrix

The new Solution Package for TIBCO ActiveMatrix  provides deep visibility into the SOA development platform provided by TIBCO.  When organizations have invested in service oriented development and delivery environments, this piece becomes the master of the universe.  Once people got over the “reusable services” buzz around SOA, they quickly got to the practical concerns of such an architecture.  With published services offered to any application team to re-use at will, you need to have high visibility on the load of those services and whether they are performing well.

One key capability of this Solution Package is the ability to view all nodes that are supporting a given service. TIBCO provides the capability to balance the load of various services into multiple nodes for scalability and fault-tolerance. However, the onus is on the support teams to figure out if it is configured correctly and performing efficiently to ensure that the load is balanced equally while in the flux of daily operations.  The Solution Package for TIBCO Active Matrix provides the visibility that these support teams require to support what is now such a crucial part of the application infrastructure.

UX Monitor Solution Package

The UX Monitor Solution Package provides the means to determine how an end-user would experience the performance of your web applications. It does so by performing simulated transactions against that web app and reporting how fast those transactions perform, as well as reporting any error conditions.  This information can help prioritize events that are happening in your data center by telling you whether a given infrastructure alert is impacting the user experience or not and to what level.  This information might also be valuable to help prioritize which areas of your service delivery are in need of additional investment to provide the desired service levels.

RTView Host Agent

The RTView Host Agent is an agent that collects health and performance metrics from the physical server it is deployed upon.  While many organizations will already have a solution that can monitor the health of physical servers (like TIBCO Hawk or Oracle Enterprise Manager), RTView EM is flexible enough to incorporate those metrics and alerts from existing systems.  This is consistent with our agent-less approach towards deploying monitoring solutions.

However, there are some situations where using our RTView Host Agent is preferable.  Those reasons often involve:

• Filling a gap in coverage,
• more in-depth metrics are required,
• and most commonly, a more granular frequency of data updates required to do proper analysis.

The RTView Host agent can help to fill those gaps.

For example, the RTView Host Agent can be used in conjunction with our Oracle Coherence monitor to deeply correlate how the data grid performance is related to physical server CPU, memory and network resources that are hosting the nodes that comprise the grid.  Gathering this information at 15-minute intervals provided by some monitoring systems is useless for this purpose, whereas the RTView Host Agent can be configured to get this information in seconds.

Oracle Enterprise Manager Connector

As discussed previously, we approach the whole end-to-end RTView EM monitoring solution as added value to your existing siloed monitoring systems. End-to-end monitoring solutions often promote this capability but then the integration and maintenance of those systems are left to the user to manage.

The OEM Connector provides a simple and supported way to expose the performance metrics of databases and physical servers that have been configured as monitored targets in OEM.  Once you do that, you get the out-of-the-box deep visualization and analysis of those performance metrics in RTView EM as well as the correlation of those metrics and how they affect your applications.

In that scenario, OEM is still used for the daily management of Oracle resources, and RTView EM is used by support teams to determine the health of those Oracle components in relation to your deployed applications and your heterogeneous application stack.

Metric Explorer

One of the main unique benefits of RTView Enterprise Monitor is that it includes the RTView Builder.  The RTView Builder has been used for over a decade to provide the means for creating very rich interactive dashboard interfaces to real-time data.  Because of this feature, there is almost no limit to customizing the end-user interface that you want to create to help support teams perform their jobs efficiently and within your organization’s support workflow.

However, there is the use case where end users want to create very simple custom dashboards in seconds to analyze metrics that are critical to performance analysis.  The Metric Explorer feature is designed to meet this use case.

Users can quickly select any collected performance metric and construct views to analyze and correlate the historical performance of those metrics.  Those views can then be indicated as “private” or shared with the other members of the support team.  With this feature, each end-user is empowered to quickly construct views of data that are critical either to solve the problem of the day, or to expose common pain points that are critical to the daily general health of the application.

—————–

With this release we have incorporated actionable data from a number of new data sources as well as provided you options on the preferred way of collecting that data.  The Metric Explorer is our first step in what will eventually have many more features, that make it really easy to create analytic dashboards for people of all skill sets and roles in your organization.

 If you have questions and comments, we’d love to hear them.  Just leave us a comment below.