How to Get the Most Out of Network Performance Management Tools?

EMA has studied how organizations use multiple network performance management (NPM) tools and how they can integrate them to improve efficiency. In this article, Shamus McGillicuddy, EMA's director of network management research, presents findings from the "Network Performance Management in Today's Digital Enterprise" study, which surveyed 250 network managers and offers advice on how to deal with this issue.

[[272836]]

It is common for IT organizations to install three to six network performance management (NPM) tools. If they remain isolated, network operations will be fragmented and inefficient. This is an ongoing challenge for network administrators.

EMA asked 250 network managers to identify their preferred purchasing strategy for NPM tools and found that enterprises have a strong preference for fully integrated, multi-functional platforms. EMA found that enterprises that currently have 11 or more NPM tools are most likely to express a preference for this fully integrated strategy. So, while they express a desire to consolidate tools, they are not succeeding.

Why use multiple NPM tools?

Part of the problem is that enterprises use NPM tools to collect and analyze so many different types of data. Infrastructure metrics collected through SNMP MIBs and traps are fundamental data sources for NPM, but they do not contain the answers that network administrators need from NPM tools.

Most organizations also collect flow data, including network flows, packets, or both. The EMA study also found that there is a strong interest in synthetic traffic generated by active monitoring tools. The most popular data source for NPM analysis is the management system API. In other words, network administrators are very interested in pulling data from other IT management systems into NPM tools for contextual analysis.

Given this diversity of data, tool fragmentation is inevitable. After all, no single vendor excels at collecting and analyzing every type of data mentioned above. They typically excel at one or two types of data, which means that enterprises inevitably need additional NPM tools to fill in gaps in visibility.

NPM tools’ connections

EMA asked survey participants to reveal how they correlate insights from multiple NPM tools. The most popular approach (25% of respondents) was to use a network operations management platform or integration between managers to gain insights from multiple NPM tools. These platforms typically excel at event management and alert correlation across multiple NPM sources.

Next, 19% mentioned direct integration between point tools, so that one tool can correlate insights captured in another tool. This approach can become complicated if an enterprise uses more than two tools.

Another 19% integrate their NPM tools with AI-enabled IT operations (AIOps) advanced IT analytics platforms, 15% integrate NPM tools with service management platforms, and 14% stream NPM data to a data lake for correlation. Only 7% claim to perform these correlations manually. That’s good, because it’s an inefficient and error-prone approach. A minority claim they have no correlations between their various tools.

EMA also asked companies how successful they were in this cross-tool linkage. 27% said they were very successful, and 49% were successful. The rest were somewhat successful, somewhat unsuccessful, or unsure. EMA categorized this 24% as “not very successful.” This question about success allowed EMA to look for potential best practices.

Companies that manually correlate insights between tools tend to fall into the “less successful” category. The most common correlation methods — direct integration between tools and integration with managers — did not have a statistically significant correlation with success.

Best Practices

Three less popular cross-tool correlation methods preferred by successful businesses.

Successful enterprises prefer to integrate with service management platforms or correlate by streaming NPM data into a data lake for analysis. Very successful enterprises integrate their NPM tools with AIOps platforms.

EMA believes that the latter three approaches are best practices for addressing the sprawl of NPM tools. AIOps tools appear to be good options.

EMA recommends that enterprises investigate AIOps platforms if they are struggling with network management tool sprawl. However, integration with a service management platform or using a data lake with a data analytics stack may also help.