Why Rely on Blind AIMDs?

We are motivated by the fact that fixed Increase rates and Decrease ratios for AIMD cannot adjust TCP’s performance to the Internet’s diverse networking conditions. Indeed, we find that fixed values for the increase/decrease factors of AIMD restrain flexibility, which is a fundamental property of transport protocols in order to guarantee utilisation and fairness in Modern and Future Internetworks. We propose a new paradigm for AIMD designs that has the potential to adjust TCP’s behaviour according to network conditions. The proposed Multi-Rate AIMD increases additively the Additive Increase factor of AIMD when the level of network contention is low and decreases multiplicatively (the AI factor) on the face of triple duplicate ACKs or timeout expirations. The Multiplicative Decrease factor of AIMD remains constant in order to guarantee fairness and stability. We show that MR-AIMD reduces retransmission effort significantly, when contention is high, becomes aggressive when contention decreases and tolerates against random, transient errors due to fading channels.