Stateless Active Queue Management Research Articles

DeepQoS: Core-Stateless Hierarchical QoS in Programmable Switches

Novel applications and network scenarios challenge existing traffic management strategies. Hierarchical Quality of Service (HQoS) provides a fine control of resource sharing and delay, but traditional HQoS solutions have challenging complexity that prevents their deployment in the traffic management engine of high-speed switches. Programmable switches have emerged to make the packet processing pipelines flexible and reconfigurable, but their traffic management capabilities still rely on fixed functions that cannot handle the complexity of traditional HQoS approaches. In this paper, we show how the extended programmability can help in addressing this challenge by the application of a fundamentally different algorithm. To emulate HQoS behavior we extend our core-stateless resource sharing framework called Per Packet Value (PPV) with a HQoS-packet marker architecture called DeepQoS. DeepQoS can be used to mark resource sharing policies of different layers simultaneously and effectively at a single point, e.g., ensuring the fair share of a household’s traffic within an access aggregation network, while also controlling the shares of its subflows. In the PPV framework, bottleneck scheduling is very simple and is unaware of flows and policies, which are encoded to Packet Values. DeepQoS can use these existing simple PPV schedulers without any change. To demonstrate the deployability of the proposed method, we have created the DeepQoS marker implementation in DPDK while redesigned and implemented our packet scheduler called Virtual Dual Queue Core Stateless Active Queue Management (VDQ-CSAQM) on a P4-programmable switch. Using extensive measurements we demonstrate the unique capabilities of DeepQoS to realize rich and deep HQoS.

A stateless fairness-driven active queue management scheme for efficient and fair bandwidth allocation in congested Internet routers

Fair bandwidth sharing is important for the Internet architecture to be more accommodative of the heterogeneity. The Internet relies primarily on the end-systems to cooperatively deploy congestion control mechanisms for achieving high network utilization and some degree of fairness among flows. However, the cooperative behavior may be abandoned by some end-systems that act selfishly to be more competitive through bandwidth abuse. The result can be severe unfairness and even congestion collapse. Fairness-driven active queue management, thus, becomes essential for allocating the shared bottleneck bandwidth fairly among competing flows. This paper proposes a novel stateless active queue management algorithm, termed CHOKeH, to enforce fairness in bottleneck routers. CHOKeH splits the queue into dynamic regions at each packet arrival and treats each region differently for performing matched-drops using a dynamically updated drawing factor, which is based on the level of queue occupancy and the buffer size. In this way, CHOKeH can effectively identify and restrict unfair flows from dominating the bandwidth by discarding more packets from these flows. The performance of CHOKeH is studied through extensive simulations. The results demonstrate that CHOKeH is well suited for fair bandwidth allocation even in the presence of multiple unresponsive flows and across a wider range of buffer sizes. The results also show the ability of CHOKeH to provide inter-protocol and intra-protocols fairness and protection for short-lived flows. With a low per-packet-processing complexity, CHOKeH is amenable to implementation in core routers to offer an effective incentive structure for end-systems to self-impose some form of congestion control.

On the Transient Behavior of CHOKe

CHOKe is a simple and stateless active queue management (AQM) scheme. Apart from low operational overhead, a highly attractive property of CHOKe is that it can protect responsive TCP flows from unresponsive UDP flows. Particularly, previous works have proven that CHOKe is able to bound both bandwidth share and buffer share of (a possible aggregate) UDP traffic (flow) on a link. However, these studies consider, and pertain only to, a steady state where the queue reaches equilibrium in the presence of many (long-lived) TCP flows and an unresponsive UDP flow of fixed arrival rate. If the steady state conditions are perturbed, particularly when UDP traffic rate changes over time, it is unclear whether the protection property of CHOKe still holds. Indeed, it can be examined, for example, that when UDP rate suddenly becomes 0 (i.e., flow stops), the unresponsive flow may assume close to full utilization in sub-RTT scales, potentially starving out the TCP flows. To explain this apparent discrepancy, this paper investigates CHOKe queue properties in a transient regime, which is the time period of transition between two steady states of the queue, initiated when the rate of the unresponsive flow changes. Explicit expressions that characterize flow throughputs in transient regimes are derived. These results provide additional understanding of CHOKe, and give some explanation on its intriguing behavior in the transient regime.

Generalizing the CHOKe flow protection

This paper proposes a simple and stateless active queue management (AQM) scheme, called geometric CHOKe (gCHOKe), to protect responsive flows from unresponsive ones. gCHOKe has its root in and is a generalization of the original CHOKe. It provides an additional power of protection, achieved by introducing an extra flow matching trial following each successful flow comparison of packets. The maximum number of comparisons permitted for an arrival can be controlled by a parameter called maxcomp. The quality of flow protection improves with maxcomp. Compared to the plain CHOKe (which is just the simplest case of gCHOKe), our analysis and simulations show that the scheme can achieve over 20% improvement in the bounds of both bandwidth and buffer space used by an aggressive flow. In addition, up to 14% of the total link capacity can be saved from the unresponsive flow, allowing responsive or rate-adaptive flows to obtain a better share of resources in the router.

A queue management algorithm for fair bandwidth allocation

In this paper, a stateless active queue management algorithm called “PUNSI” which aims to achieve fair bandwidth allocation in congested routers is proposed. This algorithm prevents unresponsive flows from dominating available bandwidth shared with responsive flows by penalizing packets from unresponsive flows with a higher probability than those from responsive flows. It is motivated by the observation that unresponsive flows tend to generate traffic of higher rates than responsive ones and that, when a packet is dropped due to buffer overflow, fellow packets from the same flow seem to be found in the buffer among those having joined recently. Simulation experiments show that the algorithm discards packets from unresponsive flows effectively and achieves pretty good fair share among flows with low implementation complexity.