Ideal Characteristics of a Good Mainframe Network Monitor

Timely and comprehensive monitoring is crucial for mainframe management. When mainframe network monitoring happens in real time, it is easier to track any network problems and fix them before they turn into outages.

Real-time monitoring
A network monitor helps improve network performance only when it reports the status and performance of a network in real time, without itself being a drag on network performance. A network monitor should get data frequently enough to provide immediate alerts and real-time reports, yet run efficiently enough that it does not interfere with performance of the network.

Consumes lesser CPU resources
An ideal network monitor consumes little of CPU capacity. A monitor that consumes 1% or less of CPU capacity is much more helpful than one that needs 5% of CPU. Consuming less CPU capacity also means quicker providing of real-time alerts and reports.

In order to use less CPU resources, a good mainframe network monitor should follow efficient data-collection architecture, like getting MIB data without SNMP requests, measuring stack traffic without constant packet tracing, etc.

Enables central view of the entire network
Another characteristic of a truly useful network monitor is that it provides a central view of all of the network’s z/OS hosts, TCP/IP stacks, applications, and connections. Thus it presents a single, coherent image of system health and configuration. It provides graphic summary reports that show the health of multiple TCP/IP stacks on multiple z/OS systems. It also creates intuitive links to TCP/IP resources such as stacks, interfaces, applications and connections and also to z/OS resources such as Enterprise Extender, OSA, Sysplex Distributor, and CSM buffers.

Easy to use interface
A good network monitor uses a browser-based GUI to provide simple and intuitive navigation to any detail. It monitors all the stacks, interfaces, TCP and SNA applications, routers, and buffers from a single screen, and summarizes system health, critical statistics, and alerts in a single image. But it also lets users easily drill down to individual applications, interfaces, and stacks, and to details regarding every individual resource and event on the network.

Browser-based tool for using system operator commands
Finally, an ideal network monitor provides easy-to-use, browser-based graphic tools for network diagnosis, management and repairs. In case of a TCP/IP problem, users can easily diagnose and take immediate corrective action; the intuitive interface guides them through the process. A graphic, on-line IP packet trace facility provides an extremely powerful problem-solving tool. When the monitor suspects a security breach, its operator command interface makes it simple to query the intrusion detection systems at every z/OS image from single browser screen.

Network performance and availability are important for business. Thus a good network monitor is crucial for maintaining your mainframe network without outages, and without slowing network performance in the process.