Monitoring from Backbone Nodes vs. Cloud Nodes

User Experiences

For today’s tech tip, we’re going to pivot away from our line items, and focus on the underpinnings of Catchpoint’s Monitoring Solution. The over 800 monitoring nodes we have positioned around the world include a wide range of node types, including backbone and broadband ISPs, cloud providers, mobile, and last mile, which, along with enterprise nodes, allow us to monitor the entire service delivery chain.

When Should You Use Cloud/Backbone Nodes?

When you’re building a synthetic test, you need to decide what type of node you want to run it on. These nodes are standing in for your user and while you’ll find every node type can provide detailed telemetry, you’ll also find critical differences based on where the node is hosted and what information you can best gather from it.

In this tip, we will focus on backbone nodes and cloud nodes (two of the most common nodes used to power synthetic monitoring), providing you with a quick reference guide to help you use both node types more effectively.

When to Use Backbone and Broadband Nodes

Backbone and broadband nodes are deployed on high-speed lines inside Tier-1 ISPs, such as Comcast, Verizon, and Level3.

When should they be used?

  • To monitor overall experience instead of individual requests
  • Baselining performance data collection
  • Competitive benchmarking
  • To measure SLAs
  • Evaluating performance on a geographic basis
  • CDN monitoring
  • To monitor ISP-level peering and connectivity
  • To avoid cloud-related routing and bot-detection challenges
  • Alerting for performance and availability issues

When to Use Cloud Nodes

Cloud nodes allow you to monitor from the cloud provider itself. When can they be best used?

  • Cloud (usually) executes faster for cloud applications
  • To verify your SLAs with your cloud provider
  • To perform first mile testing of your infrastructure
  • For rapid/repeated testing in a short time span
  • Comparing Monitoring from Backbone and Cloud Nodes

The primary reason to use cloud nodes is that they allow you to evaluate how well your cloud provider is serving you. If there’s reason to doubt how available or reachable your cloud-hosted application is, running a synthetic test to your own network from the cloud helps expand the surface area your nodes are looking at – and in a sense, helps you identify all the pieces of the puzzle when something goes wrong.

Cloud nodes are likewise better for scenarios where you need to run a large volume of tests, as cloud tests are less expensive point-wise than backbone tests. However, it is worth considering the fact that cloud nodes reduce the scope of what’s being monitored, meaning the data becomes applicable to fewer scenarios. Backbone nodes are inherently more reliable than cloud nodes due to built-in redundancies and their relationship to the end-user experience.

Additionally, monitoring from the cloud adds new variables which can create noise relative to the useful data. There can also be issues related to bot-detection, tier-1 ISP backbone nodes tend not to have this issue. Bear in mind that the best monitoring solution is the one that offers its users a large variety of node types to tailor to their specific use cases.

In this week’s video, we will learn about backbone nodes vs. cloud nodes, allowing you to:

  • Gain a holistic understanding of the purpose of synthetic monitoring;
  • Understand the role of backbone and cloud nodes and how they’re set up;
  • Find out more about the pros and cons of using backbone and cloud nodes for synthetic testing; and
  • Gain insight into example scenarios that are best suited to backbone and cloud nodes.

Published on
Dec 01, 2020
Back To Top