# Paper review: Analysis and Design of an Adaptive Virtual Queue Algorithm for Active Queue Management

## Reviewer: Mike Liu

1. State the problem the paper is trying to solve.
2. The main problem the paper is trying to solve is how to design and analyze a virtual queue-based marking scheme for Active Queue Management in Internet Routers.
3. State the main contribution of the paper: solving a new problem, proposing a new algorithm, or presenting a new evaluation (analysis). If a new problem, why was the problem important? Is the problem still important today? Will the problem be important tomorrow?  If a new algorithm or new evaluation (analysis), what are the improvements over previous algorithms or evaluations? How do they come up with the new algorithm or evaluation?
4. The main contribution of this paper is that it presents a new Active Queue management scheme called Adaptive Virtual Queue (AVQ). It also presents an experimental and a mathematically strong argument for why the scheme is stable in the presence of feedback delays, is able to maintain small queue lengths, and is robust in the presence of extremely short flow when parameters are design using their presented rules. The problem is still relevant today as we look for ways to maintain stability in the face of growing traffic and congestions on the Internet.
5. Summarize the (at most) 3 key main ideas (each in 1 sentence.)
6. The three 3 key main ideas are: (1) The authors present a new scheme for Active Queue Management called Adaptive Virtual Queue and study its following properties: stability in the presence of feedback delays,its ability to maintain small queue lengths and its robustness in the presence of extremely short flows. (2) The authors present a simple rule based on control theory-defined stability to design the parametes fo the AVQ algorithm. (3) The authors compare the performance of the algorithm through simulation with several well-known AQM schemes such as RED, REM, PI controller and a non-adaptive virtual queue algorithm.
7. Critique the main contribution
• Rate the significance of the paper on a scale of 5 (breakthrough), 4 (significant contribution), 3 (modest contribution), 2 (incremental contribution), 1 (no contribution or negative contribution). Explain your rating in a sentence or two.
• I give this paper a rating of 4 because it presents significant analysis of Active Queue Management using modern Control Theory and its ideas about stability. This is a novel acheivement since control theory is a very appropriate analysis tool given that congestion on the internet is concerned with stability and speed of stability.
• Rate how convincing the methodology is: how do the authors justify the solution approach or evaluation? Do the authors use arguments, analyses, experiments, simulations, or a combination of them? Do the claims and conclusions follow from the arguments, analyses or experiments? Are the assumptions realistic (at the time of the research)? Are the assumptions still valid today? Are the experiments well designed? Are there different experiments that would be more convincing? Are there other alternatives the authors should have considered? (And, of course, is the paper free of smethodological errors.)
• The authors' methodology was rather convincing in a mathematical sense. The authors defined their algorithm using a linearized model of their non-linear TCP/AQM model and then did a state-space analysis of the effects of changing it various parameters. The authors performed a number of experiments to try to prove that there algorithm was effective at maintaining stability, small queues, and robustness in the presence of small flows, as compared to other algorithms that have previous been presented.
• What is the most important limitation of the approach?
• The most important limitation of such an approch is that it can be very convoluted to most readers and can often be so elusive that is both difficult to understand and also to critique and evaluate for errors. The limitation of their experimental approach was that there can be question to how accurately their experimental models simulate the real internet and the performance of all the algorithms on the real internet.
8. What lessons should researchers and builders take away from this work. What (if any) questions does this work leave open?
9. The lessons that researchers should take away from this work are that it is possible to do a Control Theory/State-space analysis of the Internet and its congestion algorithms given the correct model and the correct representation of the problem. Using Control Theory/State-Space analysis can yield useful information and insight into the performace and limitations of proposed algorithms.