How to Gain Visibility into Internet Performance

Continued cloud adoption is leading to an increasing reliance on internet services, and on a complex mix of external service providers and technologies to deliver those services.

For network operations teams, these moves significantly reduce visibility into the performance of the underlying infrastructure that business services depend upon. In spite of this diminishing visibility and control, these teams remain responsible for network performance.

Traditional monitoring methods gather passive device-level data, but these approaches are not possible when teams do not own or manage the network. Network operations teams must establish new techniques that help them gain the end-to-end internet visibility they need, so they can isolate issues more efficiently and more consistently avoid disruption. In order to accomplish this goal monitoring solutions should provide internet visibility in the following ways.

Fully map the route user traffic takes

When things go wrong, it is critical to identify the problem’s domain and who’s responsible for solving it—even if the issue arises outside of the team’s internal networks. The increased adoption of new technologies makes isolating issues difficult. Narrowing the scope is the first step.

By segmenting network traffic into specific domains of ownership in which issues can occur, network operations teams can more readily identify who is at fault. Being able to gain visibility into these error domains is essential to reducing the mean time to resolution (MTTR) of issues.

The core error domains to focus on covering with exercise of mapping the full route are the end-user domain, last-mile ISP domain, mid-path or transit domain, and the application service provider domain. Each of these requires slightly different insight and is more often owned by different external parties. By continuously determining the dynamic route between locations or between users and applications network operations can build a better understanding of where traffic is flowing through the day, week, or month.

Continuously monitor performance over the route

On a 24-7 basis, test and consume application and network resources, just like users do. In this way, teams can catch classes of problems that go unnoticed by passive device monitoring, such as protocol and router configuration problems. A core challenge with using route analysis to gain actionable visibility into internet performance is the fact that the routes between business-critical apps and users are dynamic. At any given time, multiple third-party networks, diverse load balancing systems, security tooling and dynamic routing technologies may be in play.

Traditional passive monitoring based on packet and flow data can be used to understand the capacity needs for an office or to set a minimum baseline for remote employees. Active monitoring is key to understanding the continuous performance of links and to acquiring the insights needed to speed issue resolution.

AppNeta by Broadcom features TruPath technology that delivers the active monitoring capabilities that today’s network operations teams need. The solution leverages common protocols to continuously monitor network paths. When thresholds are exceeded, the solution provides advanced escalation of monitoring to gather more diagnostic data.

Understand the full responsibility footprint of applications and networks

Teams need to monitor as much of the organization’s network footprint as possible. Put low-overhead monitoring on every network that users rely on and on important infrastructure. This is essential in determining who is actually affected when problems occur. For many organizations the challenge has to do with scale. Metrics gathered across multi-technology, multi-vendor environments and unmanaged networks should be leveraged in a unified fashion, and delivered via reporting, dashboards, and APIs.

Solutions for monitoring and observability should provide insights into how much network capacity is being used and where traffic is flowing. Techniques like advanced baseline monitoring and threshold functions should be employed, including capabilities for tracking deviation from normal and generating alarms based on time over thresholds.

Finally, ISP validation should be a core focus for organizations using direct-to-internet routing, even if those routes are merely an underlay for an SD-WAN deployment.

Bringing it all together

Organizations continue to grow increasingly reliant upon internet connectivity for even the most basic business functions. This means that network operations teams will remain accountable for the performance of networks they don’t natively have visibility into. While device-level infrastructure data remains important, it is not enough to tackle modern network issues that involve growing enterprise sites and footprints.

Given these realities, it is now incumbent on network operations teams to fully map the route traffic takes, continuously monitor performance over critical routes, and understand the full footprint of networks and applications.

Broadcom helps organizations with internet visibility in order to enhance connected experiences for all internal and external end users. Explore these ways in more depth by reading our complimentary white paper.