Designing Run-Time Environments to Have Predefined Global Dynamics

TL;DR

This paper introduces a novel approach to designing run-time environments for networking protocols that ensure predictable global dynamics, combining design-time analysis with run-time adaptation for improved stability and performance.

Contribution

It presents a combined design-time and run-time framework for creating self-optimizing network protocols with predictable behavior, validated through simulations and real-world tests.

Findings

Feasibility demonstrated in simulations.

Effective run-time adaptation in real-world setups.

Enhanced stability and predictability of network protocols.

Abstract

The stability and the predictability of a computer network algorithm's performance are as important as the main functional purpose of networking software. However, asserting or deriving such properties from the finite state machine implementations of protocols is hard and, except for singular cases like TCP, is not done today. In this paper, we propose to design and study run-time environments for networking protocols which inherently enforce desirable, predictable global dynamics. To this end we merge two complementary design approaches: (i) A design-time and bottom up approach that enables us to engineer algorithms based on an analyzable (reaction) flow model. (ii) A run-time and top-down approach based on an autonomous stack composition framework, which switches among implementation alternatives to find optimal operation configurations. We demonstrate the feasibility of our self-optimizing system in both simulations and real-world Internet setups.