The correct way to calculate network bandwidth requirements

Different networks have different bandwidth requirements. How to correctly calculate the bandwidth is the key to creating and maintaining a stable and fast network.

Most network administrators know that bandwidth is an important factor to consider when designing and maintaining a stable LAN or WAN. Unlike servers, which can be repeatedly configured during the life cycle of the network, bandwidth is usually determined at the beginning of network design using the correct bandwidth formula.

[[194465]]

Want to know how to calculate bandwidth requirements when designing a network? What are the things to pay attention to in this process? This article will answer these questions one by one.

Understanding Bandwidth

Bandwidth refers to the data transfer rate supported by a network connection or the network card connected to it. It refers to both capacity and time, and refers to the amount of data that can be transferred between two points in a specific time interval. It is usually measured in bits per second (bps), and sometimes bytes per second (bps).

Network bandwidth represents the capacity of a network connection, but when determining the network bandwidth formula, it is important to understand the difference between theoretical throughput and actual throughput. For example, a 1000BASE-T (using unshielded twisted pair) Gigabit Ethernet (GbE) network can theoretically support a bandwidth of 1,000 megabits per second (Mbps), but this speed cannot be achieved in practice due to hardware and system software overload issues.

One thing to keep in mind when thinking about how to calculate network bandwidth requirements is that bandwidth should not be confused with throughput, which refers to speed. While high-bandwidth networks are generally faster, there are exceptions. A good analogy for thinking about bandwidth is cars on a highway. A high-bandwidth network is like a six-lane highway that can pass hundreds of cars at any given time. A low-bandwidth network is like a one-way street where all cars have to queue up to pass.

While cars on a large highway may be able to move faster, traffic may grind to a halt. Or it may be the case that at least small cars cannot move fast on a highway because a few large trucks take up most of the road. Similarly, even a high-bandwidth network may slow down due to problems such as congestion and high-bandwidth applications.

This is the problem that makes bandwidth requirement calculations challenging, but bandwidth calculation formulas can also cause problems. If there is not enough bandwidth, then the network will inevitably be slow. However, the cost of severely over-allocating bandwidth is something most enterprises cannot afford.

So how do you determine the right formula to meet your bandwidth requirements? The process starts with identifying these questions: What applications are users using, and what are the performance service level agreements for those applications? I know that some network managers only care about how many users are coming from a virtual LAN. But what you really need to know is what those users will be doing on the network. It's possible that 200 users will cause less network congestion than 3 users. For example, if they are using some terrible client-server application or overusing high-bandwidth services, such as high-definition video conferencing.

Bandwidth calculation formula

Calculating bandwidth requirements involves two steps:

1. Determine the amount of available network bandwidth.

2. Determine the average usage required for a specific application.

Both values ​​can be expressed in bytes per second. Consider this formula: A GbE network has 125,000,000 Bps of available bandwidth. This is calculated by dividing the number of bits (10 billion for a GbE network) by 8 to get the number of bytes.

After determining the network bandwidth, you need to calculate the amount of bandwidth used by each application. Use a network analyzer to determine the number of bytes per second that the application sends in network packets. To do this, first enable the Cumulative Bytes column on the network analyzer. The subsequent calculations for the bandwidth formula are:

1. Capture traffic to and from the test workstation running the application.

2. In the Decode Summary window, mark the initial packet of the file transfer.

3. Find the time cutoff after 1 second, and then query the cumulative bytes field at that time.

If you find that the amount of data transmitted by the application is 200,000 Bps, you can calculate it like this: 125,000,000 Bps ÷ 200,000 = 625 concurrent users. In this way, even if there are hundreds of concurrent users connected to the network, the network will not have any problems.

However, if the network is 100Mbps, the calculation formula becomes: 13,102,000 Bps ÷ 200,000. Then, this network can support no more than 60 concurrent users running applications at the same time. Therefore, knowing the formula for calculating bandwidth is important knowledge for network administrators.

***One suggestion: capture 10 seconds of data and then do the division. Also, ***check multiple workstations to ensure that the final result truly reflects the overall situation of the network. Also, be sure to determine the number of concurrent users on the network.