Poor network performance doesn't stay contained to one call. A dropped packet here, a spike in jitter there, and an entire unified communications deployment starts to feel unreliable — even when the platform itself is working exactly as designed.
Large and small organizations alike are no longer tied to a fixed office or a wired connection to stay in touch. But sub-par network connectivity and poor call quality are major issues for many businesses, and network problems can seriously disrupt a communication and collaboration framework. That's what makes dedicated call quality testing tools — and a network quality manager like IR Collaborate — essential.
How can you optimize your VoIP network's performance? Read our guide
Beyond the call itself, Unified Communications (UC) ecosystems bring together many different tools, devices, and applications — all of which depend on a solid network infrastructure to work as one system.
Vital components of a resilient network include:
One of the most common issues IT professionals face is packet loss, which can significantly impact network performance. The right packet loss test tool can accurately measure packet loss, identify Voice over Internet Protocol (VoIP) performance problems, and pinpoint issues with video calls.
In this article, we'll explain why it's vital to have a feature-rich packet loss test tool. With the right solutions, you can run packet loss tests to catch these issues early — before they degrade network performance and start affecting your organization's bottom line.
Packet loss happens when data never reaches its destination — and it's usually the reason your calls sound choppy.
Network packet loss occurs when data packets being transmitted over an internet connection are lost or dropped in transit and never reach their destination, leaving gaps in the information.
Lost packets, or delayed data transmission, can seriously affect network and application performance — not just the user experience, but business operations more broadly.
The best way to describe packet loss is the "choppiness" you might hear on a call: parts of a voice or video conversation get lost in transmission, resulting in fragmented video streaming and unclear audio.
Packets are transmitted through Transmission Control Protocol (TCP), which divides a file into efficiently sized packets for routing. Each packet carries its own number and the destination's internet address, and each may travel a different route to get there. When they arrive, TCP reassembles them into the original file at the receiving end.
Packet loss is one of three major network performance issues, along with latency and jitter.
Congestion, aging hardware, security threats, and overtaxed devices all show up as the same symptom: dropped packets.
There are several causes of packet loss — and just as many ways to test for it.
The primary cause of network packet loss is congestion — network space limitations that create a bottleneck, so traffic doesn't reach its destination on time.
Congestion at peak times means packets may be discarded and have to wait to be delivered. Most software is built to automatically retrieve and resend those discarded packets, or reduce transfer speed to improve the chance of delivery.
With the rapid pace of digital transformation, hardware and software become outdated or redundant faster than ever.
Firewalls, routers, and network switches consume significant power, and can weaken network signals over time. Hardware or software updates can slip through the cracks during expansions or mergers, contributing to packet loss or connectivity outages.
Another common problem is buggy software running on a network device, which disrupts performance and prevents packets from being delivered.
High rates of packet loss can also point to a security breach, such as packet sniffing.
Hackers can infiltrate a router and instruct it to drop packets, or launch a denial-of-service attack (DoS) that floods the network with traffic, preventing legitimate users from accessing files, emails, or online accounts. Packet loss can be difficult to resolve during a full-blown security breach.
Networks operating above the capacity they were designed for can drop packets once they're overloaded. Many devices have built-in buffers that hold packets until they can be sent.
A packet loss test tells you not just that performance has degraded, but why.
Packet loss is the main culprit behind many network issues, particularly in wide area network (WAN) connectivity and Wi-Fi networks.
A packet loss test tool lets you quantify packet loss and confirm the reason behind it — for example, whether the issue comes from network connectivity itself, or from degraded quality caused by TCP or User Datagram Protocol (UDP) packet loss.
Ping, traceroute, and MTR each answer a different piece of the same question: where are packets actually being lost?
Packet loss monitoring solutions are the most effective way to control and troubleshoot packet loss issues.
Most network administrators use a dedicated network monitoring tool, or packet loss checker, to run regular tests, diagnose problems with wireless signal coverage or insufficient bandwidth, and carry out root cause analysis across the network.
For everyone else, there are two standard tools for testing packet loss:
Ping measures the round-trip time it takes data to travel between your computer and an internet destination.
Ping sends packets to a designated destination to check whether the node at the receiving end responds correctly.
The best approach is to send a large number of pings to the destination, then check for failed responses. For example, if a destination has been pinged 50 times and only 49 responses come back, packet loss can be estimated at around 2%. Packet loss above 5% is generally considered a cause for concern.
Traceroute measures the response time of every router along the path between your computer and an internet destination, and shows where a connection is slow or unresponsive.
The traceroute command is tracert destination on Windows, or traceroute destination on Mac OS X.
Traceroute identifies not just what the problem is, but where it is in your system — whether the fault sits with your internet service provider (ISP), your network, or your own equipment.
If a traceroute shows trouble with devices associated with your ISP, it may be worth reviewing your service level agreement.
My Traceroute (MTR) is a dynamic performance diagnostic tool that combines the functionality of ping and traceroute.
MTR checks the connection between the destination host and the source server — analyzing response time and packet loss along the route using Time to Live (TTL) and Internet Control Message Protocol (ICMP).
If you're having trouble connecting to a website, traceroute can help identify the source of the problem while visualizing the path traffic takes between your computer and a web server.
Manual testing tells you about one moment. Continuous monitoring tells you about all of them.
Network conditions can change from moment to moment. High uptime is critical, and any interruption to network function can quickly disrupt business operations.
An integrated performance management solution like IR Collaborate is built to monitor performance across your key network locations and help you manage problems like packet loss, latency, and jitter.
It also helps you establish and maintain VoIP deployment performance baselines, and monitor performance metrics across your whole network.
High-end monitoring solutions like IR Collaborate help you keep track of the overall health, performance, security, and functionality of your network devices.
A good user experience is essential in any unified communications and collaboration environment. By monitoring end-to-end performance and user experience metrics, administrators can pinpoint specific areas for improvement.
Want to know more about packet loss, jitter, and latency? Download our in-depth guide.