An NPM or network performance monitoring solution is a term used to describe the process of monitoring, measuring, adjusting, troubleshooting, and analyzing the quality of network performance. The terms network performance monitoring and network performance management are often used interchangeably.
An NPM solution is essential to help network operations teams proactively anticipate network congestion and outages and adjust policies, resource allocation, and flow paths accordingly.
NPMs introduce visibility metrics that measure network health and ensure that networks operate at their peak performance.
Faster Mean Time To Resolution (MTTR)
Having the enhanced visibility provided by an NPM solution allows the NOC monitoring team to quickly identify the root cause of network disruptions and address them before the impact is felt.
Instead of overpaying for additional bandwidth or equipment and circuit upgrades, an NPM can provide precise capacity planning visibility into the requirements and usage of each device and application. Operations teams tend to over-allocate resources rather than risk the impact of network congestion. NPMs remove the need for overspend with precision-based measurements that get your capacity planning expenditure down to a science.
Organizations are subject to GDPR for all EU citizens, CCPA for California, DEPA for India, and countless other data protection regulations by region or country. US government institutions are required by EL1 to hold packets for a minimum of 72 hours, up to a year, depending on the circumstance. HIPAA protects certain types of personal data from decryption. A modern NPM solution should consider these various requirements and be able apply specific rules to meet these industry regulations or handle encrypted traffic in new ways that don’t violate these regulations.
Where are NPMs used?
NPMs are used for capacity planning, minimizing network outages, and improving network connection performance. Their impact is most clearly seen in large, distributed networks with remote locations or complex architecture. Many of today’s global and national enterprises feature these networks and use NPMs to gain visibility into their multiple planes, providers, and architectures. NPMs are also commonly used by organizations dealing with sensitive data and are subject to regulations for breaches or outages. For example, financial institutions, healthcare organizations, insurance companies, and federal agencies to name a few.
How do NPMs work?
NPMs gather information from devices to monitor the quality of connections across a network. They do this by using flow collectors, packet analysis, or analysis of other forms of data like IPFIX flow, SNMP, and API to gain visibility into network health.
What is the difference between an NPM and an APM?
When comparing APM vs. NPM, APM is always a software-based monitoring solution focusing specifically on application performance. Many first-generation NPM tools were hardware-based, relying on probes and taps and only collecting one type of data associated with network infrastructure. For example, some legacy NPMs only offered packet capture for visibility into network device behavior but lacked visibility into cloud-hosted applications. This created a need for APM solutions. Today, modern NPMs can cover both software-based application monitoring and network core infrastructure monitoring, so IT teams no longer have to invest in two separate platforms.
What are the traits to look for in an NPM solution?
A modern NPM solution should check these ten boxes and then some to ensure you’re investing in a dynamic solution that will evolve with the changing demands of your network.
- Can integrate diverse data types like packet, flow, SNMP, API, and IPFIX
- Real-time analysis and reporting
- Visual Network Topology
- Modern device management, remote visibility into all network device metrics, ability to reset remotely, and auto-discover capabilities
- SD-WAN, Multi-Cloud, Hybrid-Cloud, and Public Cloud Monitoring
- Intelligent packet capture and retention
- Playback VoIP troubleshooting
- Capacity Planning
- Mapping of network and application dependencies
- Use of AI and Machine Learning