Bandwidth and throughput are two terms that are often used interchangeably when discussing business connectivity. But they aren't the same thing.
Understanding the difference matters when you're specifying connectivity for customers, comparing services or explaining why a connection advertised at a particular speed may not always deliver that speed in real-world conditions.
For ISPs and MSPs, getting to grips with bandwidth and throughput can help you make better connectivity recommendations, set realistic expectations and choose the right service for each customer's requirements.
What is bandwidth?
Bandwidth is the maximum amount of data that a network connection can theoretically transfer over a given period of time.
It's normally measured in bits per second, such as:
- Mbps – megabits per second
- Gbps – gigabits per second
- 10Gbps – 10 gigabits per second
Think of bandwidth as the capacity of a network connection.
A 1Gbps Ethernet service, for example, has considerably more available capacity than a 100Mbps service.
However, bandwidth doesn't necessarily tell you how much data is actually being transferred at any particular moment.
This is where throughput comes in.
What is throughput?
Throughput is the amount of data that is actually transferred successfully across a network over a given period of time.
While bandwidth describes the potential capacity of a connection, throughput describes its real-world performance.
For example, a customer may have a 1Gbps connection but achieve a throughput of less than 1Gbps when transferring data.
This doesn't necessarily mean there's a problem with the connection.
A range of factors can influence throughput, including:
- Network congestion
- Latency
- Packet loss
- Network equipment
- Router and firewall capabilities
- Wi-Fi performance
- Server capacity
- Application performance
- The capabilities of the destination network
- Protocol overhead
This means that a customer's connection speed and their actual data-transfer performance aren't always identical.
Bandwidth vs throughput: a simple analogy
One of the easiest ways to understand the difference is to think about a motorway.
Bandwidth is the number of lanes on the motorway.
The more lanes you have, the greater the potential capacity for vehicles.
Throughput is the number of vehicles actually getting from A to B.
Even with a six-lane motorway, traffic congestion, roadworks, accidents or restrictions can prevent vehicles from travelling at maximum capacity.
Networks work in a similar way.
A high-bandwidth connection provides greater potential capacity, but actual throughput depends on what is happening across the network and the equipment and applications using it.
Why doesn't throughput always equal bandwidth?
There are several reasons why measured throughput can be lower than the headline bandwidth of a service.
- Network congestion
When network resources are heavily utilised, congestion can affect the rate at which data can be transferred.
This can occur within a customer's own network or further along the path between the source and destination.
- Latency
Latency is the time it takes for data to travel between two points on a network.
High latency can affect the performance of certain applications and data transfers, particularly where there are many back-and-forth communications between systems.
For example, interactive applications such as cloud services, voice and video can be sensitive to latency even when significant bandwidth is available.
- Packet loss
Data travelling across a network is broken into packets.
If packets are lost, they may need to be retransmitted. This can reduce effective throughput and impact application performance.
Packet loss can be caused by congestion, faulty equipment, poor network configuration or problems elsewhere along the connection path.
- Network equipment
The customer's router, firewall, switch or other network infrastructure can become a bottleneck.
There's little value in purchasing a high-capacity connectivity service if the customer's edge equipment can't process traffic at the required rate.
This is particularly important when deploying 1Gbps, 10Gbps or higher-capacity services.
- Wi-Fi
A customer may have a 1Gbps fibre connection but experience considerably lower speeds over Wi-Fi.
In this scenario, the fibre connection isn't necessarily the limiting factor. The wireless network, access point, device or environmental conditions may be restricting performance.
This is an important distinction for MSPs when troubleshooting customer connectivity issues.
- End-to-end limitations
The internet or private network connection doesn't operate in isolation.
A customer downloading a large file isn't just dependent on their own connection. The performance of the remote server, the networks between the two endpoints and the available capacity at the destination can all affect throughput.
In other words, a 1Gbps connection doesn't guarantee that every service accessed across the internet will deliver data at 1Gbps.
Why bandwidth still matters
If throughput represents actual performance, does bandwidth really matter?
Absolutely.
Higher bandwidth provides greater capacity for handling simultaneous traffic and demanding applications.
Consider a business with:
- Cloud applications
- VoIP
- Video conferencing
- Cloud backups
- Large file transfers
- Multiple users
- Internet-based security systems
- Hosted applications
A connection with insufficient bandwidth can quickly become saturated as these services compete for available capacity.
Increasing bandwidth can provide additional headroom and reduce the likelihood of the connection becoming a bottleneck.
However, simply increasing bandwidth isn't always the answer.
If the real issue is latency, packet loss, poor LAN infrastructure or an underpowered firewall, upgrading the access circuit alone may not solve the problem.
Bandwidth, throughput and utilisation
Another important concept is utilisation.
If a customer's connection has 1Gbps of bandwidth but is consistently using 950Mbps, the circuit is operating close to its available capacity.
As demand approaches the connection's maximum capacity, there may be less headroom for traffic spikes.
This is why connectivity should be specified around the customer's current and anticipated requirements, rather than simply selecting the cheapest service that meets today's average usage.
For MSPs, monitoring traffic patterns can provide valuable insight into whether a customer needs more bandwidth or whether another part of the network is causing the performance issue.
Does more bandwidth always mean better performance?
Not necessarily.
Imagine two customers:
Customer A has a 1Gbps connection but an outdated firewall that can't process traffic efficiently.
Customer B has a 500Mbps connection with modern network equipment and a well-designed LAN.
Customer B could potentially experience better real-world performance for their particular applications, despite having less bandwidth available.
This highlights an important point:
Bandwidth is only one component of network performance.
When assessing connectivity, partners should consider the complete service rather than focusing solely on the headline speed.
Why the distinction matters when selling connectivity
For ISPs and MSPs, understanding bandwidth versus throughput is particularly useful when discussing connectivity with customers.
A customer asking for "the fastest possible connection" may actually have a different requirement.
For example, a business heavily reliant on cloud applications may benefit from:
- Sufficient bandwidth
- Low latency
- Reliable connectivity
- Strong SLAs
- Diverse network routing
- Appropriate customer-premises equipment
- Effective traffic management
- Resilient infrastructure
Similarly, a customer regularly transferring large datasets may have different requirements from a business primarily using SaaS applications and VoIP.
The right connectivity solution therefore starts with understanding how the connection will actually be used.
Where Ethernet fits in
For businesses with demanding connectivity requirements, Ethernet services can provide high-capacity, business-grade connectivity with appropriate service levels.
VeloxServ provides partners with access to a range of wholesale connectivity solutions, including Ethernet, leased lines and business broadband, helping ISPs and MSPs match connectivity to their customers' requirements.
For customers requiring high-capacity connectivity, Ethernet can provide a significant increase in available bandwidth compared with lower-capacity access services.
But again, the headline bandwidth is only part of the equation.
Partners should also consider factors such as service availability, latency, resilience, SLA requirements and the wider network architecture.
Don't overlook network resilience
There's another reason partners should look beyond bandwidth.
A customer could have an extremely high-capacity connection but still experience a significant outage if there's no resilience built into the network.
For business-critical applications, network diversity and resilient routing can be just as important as capacity.
A resilient connectivity strategy might involve diverse circuits, different physical routes, separate network infrastructure or multiple access technologies.
The objective is to reduce the risk that a single failure takes critical services offline.
Bandwidth vs throughput: the key differences
|
Bandwidth |
Throughput |
|
|
What it measures |
Maximum potential capacity |
Actual data transferred |
|
Think of it as |
Network capacity |
Real-world performance |
|
Typically measured in |
Mbps/Gbps |
Mbps/Gbps |
|
Affected by congestion? |
Not necessarily the headline capacity |
Yes |
|
Affected by packet loss? |
Not directly |
Yes |
|
Affected by equipment? |
Equipment needs to support the bandwidth |
Equipment can limit actual throughput |
|
Useful for |
Selecting appropriate capacity |
Assessing real-world performance |
How should partners explain it to customers?
A simple way to explain the difference is:
Bandwidth is how much data a connection could potentially carry. Throughput is how much data it is actually carrying successfully.
This distinction can prevent unrealistic expectations around connectivity speeds.
It also gives partners a better framework for diagnosing performance problems.
If a customer is experiencing slow application performance, don't immediately assume they need more bandwidth.
Instead, consider the bigger picture:
Is the connection saturated?
Is latency affecting the application?
Is there packet loss?
Can the customer's router or firewall handle the traffic?
Is Wi-Fi creating the bottleneck?
Is the destination service capable of delivering the expected performance?
Is the network sufficiently resilient?
Taking this end-to-end approach can lead to a much more effective connectivity solution.
The bottom line
Bandwidth tells you how much capacity a connection has. Throughput tells you how much data is actually being transferred.
They're closely related, but they're not interchangeable.
For partners, understanding the distinction is important when specifying services, troubleshooting customer performance and setting expectations around connectivity.
A high-bandwidth connection provides the capacity to support more users, applications and data. But achieving good real-world performance also depends on latency, packet loss, network equipment, congestion, application performance and the wider network infrastructure.
At VeloxServ, we understand that choosing connectivity isn't simply about picking the biggest number.
As an independent wholesale-only aggregator, we help ISPs and MSPs access a broad range of connectivity options from multiple suppliers, giving partners greater flexibility when designing solutions for their customers.
Need help finding the right connectivity solution for your customer? Talk to the VeloxServ team today.
