Agile-SD: A Linux-based TCP congestion control algorithm for supporting high-speed and short-distance networks

Abstract

Recently, high-speed and short-distance networks are widely deployed and their necessity is rapidly increasing everyday. This type of networks is used in several network applications; such as Local Area Networks (LAN) and Data Center Networks (DCN). In LANs and DCNs, high-speed and short-distance networks are commonly deployed to connect between computing and storage elements in order to provide rapid services. Indeed, the overall performance of such networks is significantly influenced by the Congestion Control Algorithm (CCA) which suffers from the problem of bandwidth under-utilization, especially if the applied buffer regime is very small. In this paper, a novel loss-based CCA tailored for high-speed and Short-Distance (SD) networks, namely Agile-SD, has been proposed. The main contribution of the proposed CCA is to implement the mechanism of agility factor. Further, intensive simulation experiments have been carried out to evaluate the performance of Agile-SD compared to Compound and Cubic which are the default CCAs of the most commonly used operating systems. The results of the simulation experiments show that the proposed CCA outperforms the compared CCAs in terms of average throughput, loss ratio and fairness, especially when a small buffer is applied. Moreover, Agile-SD shows lower sensitivity to the buffer size change and packet error rate variation which increases its efficiency.

Agility based safety growth of Slow-Start Congestion Avoidance and Control Scheme in TCP

“…TCP Veno adopts queuing delay as an index to adjust AIMD parameters for different random packet loss rates in fast-start-up [22,23]. The Agility based Agile-SD [24] is used in high speed and short distance networks. The performance of these TCP variants are good for the application scenarios for which they were originally designed.…”

“…This algorithm is a combination of Hybrid [25] and Agile-SD [24]. Main objective of this algorithm is to avoid the congestion before receiving three duplicate Acknowledgement (ACK).…”

“…The RTT independent epoch time method enhances the performance of Agile-SD [24] algorithm. In RTT dependent every transaction depending to RTT and followed time out for every segment.…”

“…This methodology used to reduce the delay, and also increase the network utilization. Agile-SD [24] increases its CWND independently from the Round Trip Time (RTT). The epoch time needed by the standard TCP, which is "RTT-dependent" [9], is the number of needed cycles time RTT, so it will equal to 80ms.…”

“…So, λi will take the value of (4, 3, 2, 1) sequentially, which will result in an epoch time equal to 41.66ms. Thus, the epoch time of Agile-SD [24] will be shrunk by around 48% from the epoch time of the standard TCP on the same network link. The epoch time of this algorithm will be shrunk by around 50% from the epoch time of the Agile-SD on the same network link calculated by interval difference of RTT.…”