How to Monitor Network Performance for Multiple Remote Sites?

General
1

Hey guys… :wave:

I am managing a network that spans multiple remote sites, and I am looking for advice on how to effectively monitor network performance across these locations.

Specifically, I have several branch offices that rely heavily on a consistent and reliable connection to our central data center and cloud services. In the past, we’ve encountered intermittent connectivity issues and slowdowns that have been difficult to diagnose. My goal is to proactively monitor these connections and quickly identify any issues that might arise.

I have already set up a few agents at some of our key locations, but I’m not entirely sure how to optimize the monitoring setup for best results. Here are a few questions I have:

  • What are the best practices for configuring NetBeez agents to monitor multiple remote sites?
  • How can I ensure that I’m getting accurate and actionable data from the agents?
  • Are there specific metrics or tests I should be focusing on for monitoring WAN performance?
  • How do you typically handle alerts and notifications for performance issues? What thresholds do you set?

I also check this: https://community.netbeez.net/t/api-requests-for-specific-speed-test-resultlooker But I have not found any solution. Could anyone guide me about this?

Thanks in advance!

Respected community member! :blush:

2

Hi @ketow31168, here are answers to your questions. I hope this helps.

  • What are the best practices for configuring NetBeez agents to monitor multiple remote sites?

When possible, start deploying one agent per location if you wish to only monitor the WAN and wired network performance and uptime. If you wish to also monitor wireless, you may need one or more WiFi agents based on your needs. One agent will be sufficient to monitor the service itself (authentication, and internet connectivity). More than one may be needed to verify the coverage area of service.

  • How can I ensure that I’m getting accurate and actionable data from the agents?

First configure appropriate targets and scheduled tests to gather the correct metrics. For instance, you may want to pick from preconfigured SaaS targets NetBeez offers, and then create some custom targets that include IP addresses of other agents (NetBeez helps select agents as destinations very easily) or internal host or services.

Then review the alert profiles assigned to these targets (edit a target once configured) and change as needed based on the SLAs that you want or have to enforce. For instance, if your network transports VoIP and videoconferencing calls, ensure that you have low latency and jitter for voice video traffic, and packet loss is zero or as low as possible.

  • Are there specific metrics or tests I should be focusing on for monitoring WAN performance?

Generally, I want to see the following network performance metrics in my data:

  1. Latency via ping
  2. Packet loss via ping
  3. DNS and HTTP success rate and timing
  4. Download and upload speed via network speed test
  5. TCP/UDP throughput between agents/locations if needed with iperf
  6. MOS and jitter estimates for cloud services
  • How do you typically handle alerts and notifications for performance issues? What thresholds do you set?

For thresholds you should consider the class of services your network support and implement appropriate ones. Do you expect a certain download, upload speed, or latency and packet loss? Adjust the alert profiles based on that.

As for the notification, NetBeez allows you to use out-of-the box integrations it offers (e.g. Slack, MS teams, ServiceNow, etc) or standard protocols such as SMTP, SNMP, and syslog.