APM Rolling Review: App Monitoring Made Really Easy

The first entry in our APM software Rolling Review is Indicative Software. Its Service Director offering, now in Version 8.0, promises a single-product approach to holistic application performance management that seeks to eliminate the complexity and ultra-high cost for multiple components that make many enterprises wary of traditional APM suites. In the Kickoff to this Rolling Review we discussed the myriad ways APM continues to vex organizations that have even a smidgeon of complexity in their application architectures. Traditional APM software is complex, costly—and required if IT is to have any hope of untangling the cause of performance slowdowns or discovering why network capacity is being gobbled by a particular application.

Fortunately for IT, newer vendors like Indicative are challenging established purveyors of management systems, like HP/Mercury, IBM and CA/Wily, by offering easier-to-use tools that are still plenty capable of monitoring enterprise environments.

For example, one area of confusion for IT groups is determining which type of data collection method is best for their enterprise. With synthetic transactions, software- and hardware-based probes, and client and application agents available, each vendor has a different philosophy. Indicative simplifies matters by relying on a combination approach, with agentless synthetic transactions at the core. This method focuses on a service-centric methodology to monitor the overall health and performance of the application. Service Director also has the ability to integrate into existing agents and data sources.A Walk in the Park

What we liked most about Indicative was how easy it was to set up and configure. To create our service model, we simply dragged and dropped three predefined templates—Microsoft SQL Server, Cisco switch, and Microsoft IIS Web Server—and provided authentication information. These templates include hundreds of pre-defined tests for different network and Web services.

We were also very impressed with Indicative's business-service approach: Instead of starting at packet or byte level, Indicative takes a top-down style and allows organizations to monitor and manage performance as a true service. While other products have this capability, Indicative does not require IT to manually correlate data in order to build service views; just drag and drop field-component templates into a logical service grouping.

The reporting interface is what we'd expect for a performance management tool. We could generate a number of real-time and historical reports showing discrete vales on events or overall service performance. Trends can be displayed as variances from the baseline data that has been collected. You can export data and use it as you see fit.

Take Your PickIndicative supported our entire application monitoring environment and also kept tabs on network services. While not tested in our review, Indicative supports a wide range of applications, such as J2EE, Microsoft .NET, Citrix, PeopleSoft and SAP. All of these are configured using the library, which includes more than 2,000 monitoring templates. If a device does not fit into a predefined template, IT may configure a generic template to monitor a nonspecific item, such as a TCP port or an SNMP get value.

Service Director easily set a baseline for and monitored our application environment. It quickly found several performance bottlenecks resulting from network congestion and was able to pinpoint the cause of an issue and display the metrics required to troubleshoot the problem in real time. Unless you're monitoring all of your applications, Indicative may not be able to pinpoint the root cause of a performance issue, but it can provide IT operations staff enough information to diagnose the problem.

Room for Improvement

Indicative does not provide a full-featured Web client, and an operator must have the Console installed to view Indicative's data. A Web-based client that allows access from any workstation would provide the ability to scale the application with minimal desktop support.

The Event List that Indicative provides to report on issues could also be developed further. For example, it would be useful if operators could filter data using a wider variety of fields, and we'd like to see the list more dynamically integrated into the Service View. Because the Service View and Event List are on the same screen, it would be nice if clicking on a service automatically filtered the event list to contain only events pertaining to that service. This capability would help IT quickly pinpoint those events impacting a service, without additional filtering.But these are certainly not deal breakers when weighed against Service Director's strong features. For example, Indicative includes an autodiscovery capability that allows IT to search for Web and network services that they may wish to monitor. The autodiscovery function not only discovers top-level services, it drills down into individual attributes. For example, a Cisco router is discovered, then the corresponding interfaces and ports. This is also true for application environments such as WebSphere, WebLogic and .NET, however, we did not test this capability.

Pricing for Service Director is based on the number of tests conducted. Indicative says a typical customer starts with around 5,000 tests, with some larger installation bases running 80,000 tests. The current average price is around $300,000 for software, maintenance and setup—although given the ease of use, setup is a fairly small portion of that cost.

Michael Biddick is a contributing editor for Network Computing and Executive Vice President of Solutions for Windward Consulting Group, a firm that helps organizations improve it operational efficiency. Write to him at [email protected].