A Study of AQM Schemes in Short Buffer Networks

The Computer networks these days are suffering from unnecessary latency and poor system performance. The culprit is bufferbloat, the existence of excessively large and frequently filled buffers inside the network. Excessively large and frequently filled buffers causes higher latency and jitter, and also reduces the overall throughput of the network. Oversize buffers has become part of routers because of the reasons like more is better mentality, the cheap cost of the memory, and varying link capacities dynamically. This situation is very much impacting the throughput of short lived flows when they are present along with long lived flows, as long lived flows will occupy the queue. This phenomenon is called Bufferbloat. The increasing bandwidth of the network interfaces raises an important question concerning the size of these buffers. Under Buffered routers lead to packet loss, thus adversely affecting application performance, while an over-buffered router causes increased latency, complexity and cost. To reduce packet losses and maintaining queues with short buffers, we have implemented 2 AQM(Active Queue management) schemes (Random Early Detection(RED) RED and Stochastic Fair Queuing(SFQ) in NS2 and compared the results in various classes of traffic along with Drop Tail(DT). We have carried out simulations with TCP as well as mixed traffic and compared the AQM schemes on the basis of throughput, packet drop percentage, end to end delay and average flow completion time (AFCT) and have shown the trade-off between the schemes in different classes of traffic.


Computer Networks

Technologies Needed


Project Structure

What we provide

  • Videos on Required technologies by Ravindrababu Ravula
  • Project Implementation Videos by Ravindrababu Ravula
  • Presentation Slides
  • Assignments with solutions
  • 12 weeks of expert guidance
  • Assistance to Complete the Documentation

Fee Structure

Fee for the Complete Project including the required technical courses is Rs. 20000.00(Please refer to the specific project to know the list of courses included)

Commencement and Validity

  • Programme commences from Jan 1, 2017
  • Validity for video lectures is 6-months from the commencement of the project
  • Expert guidance will be provided only for 12weeks from commencement of the project

Registration procedure

If you are interested in registering, you can make the payment in the following account either through net banking or at your nearest HDFC bank and email us the transaction id or scan copy of the pay-in-slip.
Account Name : Raudra Eduservices Private Limited
Account Number : 50200012182576
Account type : Current account
Bank : HDFC
After the payment is done, you can email us the screen shot or picture of transaction details or the pictures of the bank pay in slip at gate2014.ravindra@gmail.com. Once it is done, you will receive an acknowledgement mail regarding your payment status and will be given access to private lecture videos, assignments and source code from Jan 1 2017. You can watch the videos online anytime, anywhere and any number of times. Please note that the videos are not downloadable. Sharing your access or trying to sell or distribute videos is a legally punishable offense. Earlier we caught some people doing this and they were punished legally and a huge penalty was imposed on them.