Congestion Control in Computer Networks
A computer network experiences congestion when too much data is being transferred simultaneously, which slows down the network. Network congestion causes delays and perhaps data loss, just like traffic congestion on a crowded road. The network becomes "clogged," which makes it difficult for information to move freely from one location to another, when it is unable to manage all of the incoming data.
The TCP protocol is commonly used to control congestion on the current Internet (Vicisano et al., 1998). In the event that more packets are transmitted to the network than it can manage, congestion may result. One of the most important ideas in computer networks is congestion control. It describes the techniques used to guarantee efficient data flow and avoid network overload. All users can benefit from a reliable and effective network connection thanks to congestion control strategies. These methods are crucial for preserving modern networks' dependability and performance.
Congestion Control's Impact
Increased Network Stability:
By keeping the network from becoming overloaded, congestion control helps maintain network stability. It controls data flow to prevent excessive traffic from causing the network to malfunction or crash. Decreased Latency and Packet Loss: Data transmission may lag in the absence of congestion control, leading to delays and data loss. Congestion control helps manage traffic better, reducing these delays and ensuring fewer data packets are lost, making data transfer faster and the network more responsive.
Improved Throughput:
The network can make better use of its resources by avoiding congestion. This makes it possible to send more data in less time, which is crucial for managing big data sets and enabling high-speed applications.
Fair Resource Allocation:
Congestion control makes sure that users share network resources in an equitable manner. Everyone can use the bandwidth equally because no one user or application can consume all of it.
Improved User Experience:
Users benefit from faster and more seamless data flow. Applications, websites, and internet services operate more consistently and without irksome lag.
Decreased Latency and Packet Loss:
Data transmission may lag in the absence of congestion control, leading to delays and data loss. By improving traffic management and minimizing delays and data packet loss, congestion control speeds up data transfer and improves network responsiveness.
Network Congestion Collapse Mitigation:
In the absence of congestion control, an abrupt increase in data traffic can overwhelm the network, resulting in extreme congestion and nearly complete unusability. By effectively controlling traffic and averting such serious malfunctions, congestion control helps prevent this.
Types of Congestion Control Algorithm
- Leaky Bucket Algorithm
- Token Bucket Algorithm
1. Leaky Bucket Algorithm
The leaky bucket technique finds application in rate-limiting or network traffic shaping. Traffic shaping methods mostly employ a token bucket execution and a leaky bucket execution. This algorithm shapes burst traffic into a constant stream and regulates the rate at which traffic is supplied to the network. The inefficient use of available network resources is one of the drawbacks as compared to the leaky-bucket algorithm. Many network resources, including bandwidth, are not being used efficiently.
Likewise, every network interface has a leaky bucket, and the leaky bucket method consists of the following steps:
Step 1. The packet is tossed into the bucket when the host wishes to send it.
Step 2. The network interface sends packets at a steady pace since the bucket leaks at a steady pace.
Step 3. The leaking bucket transforms bursty traffic into consistent traffic. The bucket actually functions as a finite queue with a finite output rate.
2. Token Bucket Algorithm
Regardless of the bursty traffic, the leaky bucket algorithm has a strict output design at an average rate. In certain applications, the output is permitted to accelerate upon the arrival of significant bursts. This needs a more adaptable algorithm, ideally one that never loses data. As a result, network traffic shaping and rate-limiting are two applications for token bucket algorithms.
When to send traffic is determined by a control algorithm. The token display in the bucket determines this order. Tokens are in the bucket. Every token specifies a packet with a fixed size. To share a packet, tokens in the bucket must be removed. A flow to transmit traffic is displayed in the token display when tokens are displayed. If there is no token, no flow will send packets. Thus, a flow moves traffic in good tokens in the bucket until it reaches its maximum burst rate.
The following is a description of the steps in this algorithm:
Step 1. Tokens are tossed into the bucket on a regular basis.
Step 2. There is a maximum capacity for the bucket.
Step 3. A token is taken out of the bucket, and the packet is sent if there is a ready packet.
Step 4. The packet cannot be dispatched if the bucket contains no tokens.
