Features

Real-time Fault and Performance

Traverse can run tests against your applications, databases, network equipment or servers and indicate faults when the test fails or crosses a preset threshold (such as for database transaction rates, web server response times, disk space, bandwidth utilization, etc.). It can also parse for patterns in log files, receive SNMP traps, and generate alarms when a pattern matches.

In addition to detecting faults in real time, Traverse stores the collected data using real-time progressive aggregation techniques to store the performance data for extended periods of time (up to several years) with modest database size requirements (around 1GB per year of data). This historical data is then used for trend analysis, capacity planning reports and baseline reports based on statistical analysis of past data.

Traverse uses a unique distributed database and processing model to generate reports in real time from large volumes of historical data which is not available using traditional data warehousing techniques.

Flexible Service Container Views

Traverse allows users to create flexible "containers" of applications, devices or tests in order to see the end-to-end performance of a "service." For example, a "Payroll service" might have a database, a printer, and a payroll application all connected via a network router. This feature allows the user to create a "Payroll Service Container" and monitor all underlying components of that service in a single view. The status of the containers is updated in real time based on the status of its components. Additionally, these containers can be nested and one can determine service impact using the container reports. You can automatically create containers based on rules, and set the status of the container using rule logic (for redundant IT elements, etc.).

Delegated Authority User Model

Traverse has a unique delegated user model which allows multiple departments in an enterprise to each have their own "virtual management system" without being able to see each other's data, while allowing certain "administrators" to have read-only or read-write access to multiple departments. As an example, the network department, the server group, the database group and the application group can each have their own private accounts on the system, while allowing the help desk to have a read-only view across all the departments and the operations center to have a read-write view across all or some of the departments.

In a service provider environment, you can use this feature to offer managed services to customers.

Event Manager for Operation Centers

The Traverse Event Manager allows powerful and distributed filtering of syslogs, Windows events, SNMP traps and then acknowledge, suppress or delete these events using the Event Manager console. Ideal for Network Operation Center (NOC) environments, multiple operators can access the web-based interface in a distributed datacenter environment.

Notification Engine

Traverse has an extensible action and notification engine that features automatic escalation of problems over time, time of day-based notification and allows suppression of "dependent" alerts so as to prevent alarm floods. If an e-commerce service is down because an unreachable database due to an intermediate switch failing, the system can send out a single notification about the switch instead of sending a flood of alarms for everything that is unreachable. You can easily add new actions using the plugin framework.

RealView Dashboard

The RealView dashboard feature lets you create custom dashboards to view the performance of services and infrastructure. You can create multiple dashboards, each containing up to twenty components that can display and chart any metrics selected, and update in real time.

Panorama

The Panorama feature offers an interactive graphical representation of the devices in your network that are being monitored, including the status of the devices and the dependency relationships between them. Panorama offers three different topology layouts, flexible display filters, pan and zoom functionality, the ability to configure and save custom views, and the ability to add or remove device dependencies.

The previous version of Panorama was implemented as a client-side application that ran on the user's workstation, but it has been completely rewritten for Traverse 5.x to run within the web browser.

Network Flow Analysis

Traverse integrates with network flow and packet level data collection to provide seamless drill-down from system and device level monitoring to troubleshooting and analyzing using flow and packet data. This data provides details about the network traffic between hosts, enabling quick identification of impacted services, trouble areas, and problem sources.

Extensible APIs

Traverse has very powerful APIs which allow access to all components of the software. Users familiar with Perl or C can start using the API very quickly due to its familiar commands and interface. These APIs allow you to configure connections to other legacy products or custom applications.

Distribution and Scalability

The Traverse architecture is horizontally scalable and uses distributed databases and parallel processing to deliver real-time fault and performance reports. Additional reporting engines and data collectors can be added to the system as needed to scale to very large networks, and the BVE layer automatically presents a unified view across all the distributed data collectors. You can run Traverse single system for a small environment, or scale to hundreds of thousands of IT elements by deploying on multiple distributed servers.

The distributed DGE model allows Traverse to handle multiple NAT networks or firewall-protected LANs that might exist in large enterprises.