How to integrate IT infrastructure monitoring tools
When done successfully, performance monitoring tool integrations enhance administrative insight and productivity across a busy data center.
Compatibility planning and proper use of APIs simplify IT infrastructure monitoring and management tool integration projects.
Data center operators idealize the "single pane of glass" that promises ubiquitous management across the IT infrastructure. Most IT organizations are heterogeneous facilities with a mix of hardware and software, and use multiple tools for monitoring, management and system troubleshooting. These tools are not necessarily designed to work together; interoperability requires costly and time-consuming initiatives.
Set integration goals
Establish the scope of integration early on. If you go into a project looking for that vendor-promised single pane of glass, you'll be disappointed. Start with a reasonable goal: What data, views, logs or other output do you need to integrate? Resist the temptation to expand and tinker with the project scope as you discover new ways to access and import data from other tools. "Scope creep" delays and complicates integration. Instead, open a new project once the first one has been completed successfully.
Investigate each tool's source data. Knowing whether source data is available as comma separated values, a busy SQL database or some other format will make it easier to implement the right mechanism to locate and import the data into another IT infrastructure monitoring tool. Proprietary data sources are difficult to access and import without APIs or vendor-specific plug-ins.
Understand native compatibility
If your monitoring tool is designed for broad integration, it will handle many different data sources and tools. Nagios XI, for example, can gather information from systems across the enterprise through outside email alerts, simple network management protocol (SNMP) messaging, wizards, agents, components and plug-ins.
This flexibility enables a high level of integration with many tools, applications and systems in the data center. Using agents, it can monitor Windows, Solaris, Linux, AIX or other desktop and server OSes. Downloadable plug-ins grab data on applications like Apache Tomcat. Widgets like the Nagios Exchange dashlets show traffic reports or other specific information in the interface. Components offer optional extensions to handle new notifications, interface improvements and other features. One popular component is the Active Directory Integration that allows AD to act as an authentication source for Nagios.
Author Stephen J. Bigelow explains how to handle a plethora of IT systems management tooling.
Tools, such as open source Zabbix, also collect data by using agents. A native Zabbix agent runs on Linux, Windows and UNIX, collecting details on processor, memory, storage and I/O use. Zabbix natively supports text logs, so it gathers real-time Windows server information via the Windows Event Log and Windows Management Instrumentation. SNMP agents monitor almost any device on the network: uninterruptible power supply systems, printers, network attached storage arrays and so on. An Intelligent Platform Management Interface agent allows Zabbix to gather server processor temperatures, fan speeds, operating voltages and more. The tool also collects data from databases like SQL Server or MySQL, and can use custom scripts when needed.
Use API and SDK features
Software developers typically provide a rich suite of API routines that other software can access. For example, Amazon Web Services' APIs give developers precise control over Elastic Compute Cloud and Virtual Private Cloud services. Nagios supports a range of add-on APIs for file systems, scripting and protocols. Similarly, software development kits (SDKs) supply templates to use those outside services or application APIs.
Infrastructure monitoring tools that come with APIs and SDKs allow sophisticated and efficient integration between platforms. APIs also let developers apply complex rule sets to a large number of devices, which isn't practical through the regular interface. Some APIs enable two-way sharing between tools in the data center infrastructure monitoring ecosystem.
APIs and SDKs require expertise to develop and maintain any code for tool integration. The required skill set might have to come from software development staff or outside consultants.
About the author:
Stephen J. Bigelow is a senior technology editor at TechTarget, covering data center and virtualization technologies. He has acquired many CompTIA certifications in his more than two decades writing about the IT industry.
