What happens in a microsecond?

Why service providers need microsecond-level network visibility

In this blog, we look at the importance of a microsecond and putting the magnitude of a microsecond into perspective—what can happen in a microsecond and what can go wrong in a microsecond. We also discuss why microsecond-level visibility is critically important for managing end user experience, especially in the 5G age.

Blink! Blink! Blink!

At 350,000 microseconds per blink… those three blinks took just over a second.

Wow, I can blink three times in a second! That is fast! Yet I can see pixelation on a screen when data packets are being lost… could it be super vision?

My hearing must be even better! Well, an average human can detect a 100,000-microsecond delay in voice.

Super hearing too! I need to go get a super suit!

Microsecond requirements for maximum user experience

Is it really super-human? Network path changes are meant to occur in 50,000 microseconds or less, so fast that voice calls carried by data traversing that network will switch from one path to another with imperceptible changes to the average human!

When you consider that maximum one-way latency (delay) for key 5G services is only 5,000 microseconds, the blink of an eye doesn’t seem so fast anymore.

Even a hummingbird takes nearly four times longer to flap its wings (19,000 microseconds).

Next–generation networks are designed to adapt to changing conditions much faster than the blink of an eye. They even require speeds many times faster than the beat of a hummingbird’s wing to deliver the quality of experience (QoE) required for 5G services.

Unless you can truly “see” and measure the performance of next-generation networks, how can you be confident and guarantee that the quality of service (QoS) and, in-turn, the quality of experience, that is expected is actually being delivered? How can you prevent issues before they are “seen” or “heard” by your end users?

The answer to that question is ensuring that you select a performance management solution that is accurate, granular, and broad enough to capture what is really occurring in the network. You need to truly measure QoS and QoE at levels that allow for corrective actions BEFORE customers are impacted. For building a self-healing and self–managing network, microsecond visibility is not optional, it is fundamentally necessary.

Let that sink in for a … microsecond.

Accedian’s Skylight solution provides operators and enterprises with the ability to truly “see” what is occurring in the network (infrastructure and applications) at a microsecond level. Skylight’s microsecond accuracy for network performance and ability to look at 100% of packets, providing per-packet intel™ (metadata) combined with machine learning-driven analytics and visualization enables true macro visibility to micro-level detailed control of next–generation networks.