What is Latency in Networking?

When a user browses the Internet, checks their email, or performs most computer-related tasks, data is transferred back and forth between their computer and an application server. Latency is the delay before this transfer of data begins following an instruction to execute its transfer.

The Impact of Latency on Network Performance

Although we measure network latency in milliseconds, such a small delay will have a significant impact on network and application performance.

As companies increasingly rely on cloud-based solutions, their volume of network traffic increases. This increase in traffic flow, along with the unreliable nature of the Internet, can introduce additional network latency, resulting in performance degredation and an adverse impact on some latency-sensitive applications.

What Causes Network Latency?

A poorly designed or configured network are major contributors to network latency. Traffic traveling over unreliable internet connectivity is also a contributor to network latency, but is something we have little control over. 

Physical Distance

Traffic flows over networks — including the public Internet — by hopping from one router to another. Each router along the way adds some delay as it determines where the traffic should go and routes it toward its destination. The more hops that traffic has to travel through, the greater the network latency.

Network Congestion

The volume of traffic on the network can also impact the network latency. As the network becomes more congested, network devices will drop packets and force re-transmission, introducing more delays and impacting the delivery of the traffic.

Application Performance

While not related to the network infrastructure, poorly configured applications and the servers they run on will have a dramatic impact on performance and the user’s experience.

Impact of Latency on Different Applications

While all network connections have latency, some have “too much”. These are some applications where even a few milliseconds of latency can have a noticeable effect on the user experience.

Gaming and Financial Trading

In some industries, such as financial trading or gaming, network performance is critical. In a gaming context, even a small amount of lag can have a significant impact on the gameplay experience. For financial traders, slight delays in making trades can result in missed opportunities or less profitable results. For these and similar industries, network latency can be costly.

Videoconferencing

The COVID-19 pandemic drove a surge in videoconferencing that persists still today. Videoconferencing requires a constant stream of network packets carrying sound and video between all of the conference participants.

Network latency can have different impacts on videoconferencing. In some cases, latency will create lag, resulting in buffering, interruptions, and a degraded conference experience. In others, a videoconferencing system may have built-in latency management solutions that might cause video to freeze or become lower-quality if latency affects packet delivery.

SaaS Applications

Companies are increasingly dependent on cloud-based solutions including Software as a Service (SaaS) offerings. These cloud-based applications are hosted in the cloud and may be geographically distant from an organization and its users.

Depending on their purpose, latency in SaaS applications may have a significant impact on an organization and its productivity. If most or all of an employee’s work is performed using cloud-based applications, even a delay of a few milliseconds in the delivery of each packet can add up to significant wasted productivity for the organization over time.

Measuring and Monitoring Latency

Measuring network latency is useful for identifying sources of network latency and assessing its impact on the organization’s network and applications. Network latency can be measured by various tools depending on whether the goal is one-time measurement or continuous monitoring.

Ping and Traceroute

To evaluate network latency at a particular point in time, the ping and traceroute terminal applications — which are built into all major operating systems — can measure the latency of traffic to a system. Ping is commonly used to test whether or not a system is online, sending it several packets and measuring the time it takes for the responses to be received.

Traceroute goes a step further, using a series of packets to map out the route(s) that traffic takes to its destination. This can be helpful for latency monitoring since latency measurements to each hop can help to pinpoint where on the traffic’s route a delay may be occurring.

Network Monitoring Tools

Network monitoring tools — such as SolarWinds Network Performance Monitor and Auvik — are vital to maintaining visibility into the health of a network and the user experience. These tools provide ongoing monitoring and the ability to generate alerts if outages or increased latency occur. Additionally, the historical trends captured by these tools can be used to highlight the impact of poor network performance. 

Strategies to Minimize Latency

Network latency can have a significant negative impact on an organization and their users’ experience. Too much latency affects employee productivity and can render some applications completely unusable. However, companies have options for managing and reducing network latency on the corporate WAN.

Quality of Service (QoS)

Companies have a wide variety of application traffic traversing their network, and some applications are more latency-sensitive than others. Quality of Service (QoS) can help to manage network latency by ensuring that latency-sensitive or higher priority traffic is treated accordingly. QoS solutions identify priority applications and provide preferential treatment to minimize latency.

Network Infrastructure Upgrades

Poorly designed or configured networks will introduce latency and poor user experiences into networks. This is a key consideration when organizations pursue network transformation initiatives, with the objective of reducing latency and improving the user experience. These sources of network latency can be addressed via a redesign of the corporate WAN. By identifying common data flows and working to shorten the distance that traffic travels, an organization can reduce latency for these applications.

Content Delivery Networks (CDNs)

Content delivery networks (CDNs) are designed to cache static content geographically near to an organization and its users. By reducing the physical distance that traffic needs to travel and decreasing load on origin servers, CDNs can reduce network latency.

Future Trends and Technologies Addressing Latency

The growth of mobile and Internet of Things (IoT) devices makes network latency a significant problem for individuals and businesses alike. Several emerging and future technologies have the potential to provide additional improvements in network latency and better support the connected world.

5G Networks

While 5G technology has existed for several years, only recently have we seen 5G rollouts at scale. With users more dependent on mobile devices and networks, 5G has the potential to dramatically reduce network latency that they experience. 5G uses different algorithms and frequency ranges than previous mobile networks, enabling higher-bandwidth connections and supporting greater densities of mobile devices.

Edge Computing

As IoT devices become more common, increased cloud traffic and bandwidth consumption can cause network latency. Edge computing attempts to address this by moving computational resources to the network edge, closer to IoT devices and other sources of data.

These edge computing systems are designed to reduce the amount of data sent to cloud networks by pre-processing data to send only necessary data to the cloud and issue instructions to IoT devices based on their own processing. By reducing the volume of data traversing the cloud, edge computing reduces network latency.

Protocol Enhancements

Networking is an area of continuous research, and new protocols and algorithms are under development. As new protocols are created, they have the potential to reduce network latency by making networks more efficient and eliminating potential bottlenecks. For example, more efficient data compression algorithms can reduce the amount of data being sent over the network, which reduces both network latency and overall network congestion.

Cato Networks Can Help You Reduce Latency

Modern organizations are changing with distributed branch locations growing and work-from-anywhere dominating. Providing high-performing connectivity and an excellent user experience over an unreliable internet is becoming more daunting for everyone. Organizations cannot afford to limit the productivity of their workforce and require more robust solutions to deliver this.

SD-WAN solutions can address network performance issues in corporate WANs by incorporating QoS capabilities and optimally routing network traffic. Unfortunately, routing traffic over the public internet means the quality of your network’s performance depends on the quality of the internet, which is not always reliable.

Cato Networks overcame this challenge by building a global private backbone to deliver its SASE Cloud services. Combining the traffic optimization capabilities of SD-WAN with the improved performance of dedicated network infrastructure provides predictable-latency and a high-performance network that goes beyond basic SD-WAN solutions.

Cato SASE Cloud converges network and network security into a single software stack. By converging these capabilities, Cato SASE Cloud offers better performance, predictable latency, and more holistic security.

Learn more about improving your WAN performance with Cato SASE Cloud.