Giuseppe Cofano

Design and experimental evaluation of network-assisted strategies for HTTP adaptive streaming

In this paper we investigate several network-assisted streaming approaches which rely on active cooperation between video streaming applications and the network. We build a Video Control Plane which enforces Video Quality Fairness among concurrent video flows generated by heterogeneous client devices. To the purpose, a max-min fairness optimization problem is solved at run-time. We compare two approaches to actuate the optimal solution in an SDN network: the first one allocating network bandwidth slices to video flows, the second one guiding video players in the video bitrate selection. Performance is assessed through several QoE-related metrics, such as Video Quality Fairness, video quality, and switching frequency. The impact of client-side adaptation algorithms is also investigated.

Local SIP Overload Control: Controller Design and Optimization by Extremum Seeking

The session initiation protocol (SIP) is a signaling protocol for managing various types of real-time sessions between parties over an Internet Protocol (IP) network. An open issue is the control of overload situations that occur when the incoming flow of requests to a SIP server overcomes the server processing capacity. In particular, call establishment times increase due to overload, which triggers retransmissions and causes a further increase of the total incoming flow of requests. This paper proposes an overload control system for regulating both the queue length and the CPU load of the SIP server. The proposed control system is made of two proportional-integral controllers tuned by minimizing a proper cost function using the Extremum Seeking algorithm. A real working implementation of the SIP overload controller has been made in the open-source SIP server Kamailio. A performance evaluation and comparison of the proposed controller with the main proposals existing in the literature have been carried out. The results show that the proposed control system counteracts overload situations and provides a goodput close to the optimal while maintaining low call establishment delays and retransmission ratios.

A Control Architecture for Massive Adaptive Video Streaming Delivery

The optimization of massive video delivery, with the purpose of improving user engagement by delivering the maximum available quality of experience (QoE) to users, is a hot topic. In this paper we propose a Network Control Plane (NCP) for video streaming aimed at jointly maximizing users QoE and network utilization by reserving bandwidth on a per-flow basis. The NCP is placed on top of the controlled network and cooperates with distributed buffer-based adaptation techniques implemented at the client. It is specifically designed to take into account scalability and adaptivity issues. No communication with external entities involved in the delivery process, such as other networks or the streaming server, is required. We provide a reference implementation and a performance evaluation through simulations.

A Hybrid Model of the Akamai Adaptive Streaming Control System

Abstract Video streaming is becoming the application generating the largest fraction of the Internet traffic. Adaptive video streaming adds to classic video streaming the feature of dynamically adapting the video bitrate to track the time-varying network available bandwidth, avoid playback interruptions and ensure the delivery of the best video quality. In this paper we focus on the adaptive video streaming control system employed by Akamai, a major CDN operator whose video delivery system is used by several video streaming platforms, including Livestream. Differently from the typical client-side control, Akamai employs an interesting and unique hybrid client/server control architecture. Our purpose is to derive and validate a closed loop mathematical model of the control system, which turns out to be a hybrid automaton. The model is then analyzed to derive key properties which can be used to properly tune the controller parameters.

Characterizing adaptive video streaming control systems

Adaptive video streaming systems aim at providing the best user experience given the user device and the network available bandwidth. To the purpose, a controller selecting the video bitrate from a discrete set ℒ has to be designed. The control goal is to maximize the video bitrate while avoiding playback interruptions and minimizing video bitrate switches. In the literature two different approaches, which we name rate-based actuation and level-based actuation, have been explored. The first one adapts both the received rate and the video bitrate, whereas the second acts only on the video bitrate. In this paper we advocate the adoption of level-based actuation controllers and we propose a hybrid dynamical system that models the essential features of such a class of controllers. With this model we are able to derive the minimum obtainable video bitrate switching frequency which can be considered as a benchmark for any level-based actuation controller. Finally, we show how to design the video level set ℒ to obtain a performance trade-off between switching frequency and storage cost requirements at the servers. The theoretical results are validated through numerical simulation and experimental evaluation.

Local SIP Overload Control

The Session Initiation Protocol (SIP) is a signaling framework that allows two or more parties to establish, alter, and terminate various types of media sessions. The mechanism employed by the standard SIP is not effective in handling overload situations that occur when the incoming flow of requests overcomes the processing resources of the server. In this paper we present a local overload control system based on feedback control theory. The algorithm has been implemented in Kamailio (OpenSER) and a performance comparison with Ohta and Occupancy (OCC) overload control algorithms has been performed. The proposed control system efficiently counteracts overload situations providing a goodput which is close to the optimal while maintaining low call establishment delays and retransmission ratios. On the other hand, Ohta and OCC algorithms provide higher call establishment delays and retransmission ratio and lower goodputs.

Design and Performance Evaluation of Network-assisted Control Strategies for HTTP Adaptive Streaming

This article investigates several network-assisted streaming approaches that rely on active cooperation between video streaming applications and the network. We build a Video Control Plane that enforces Video Quality Fairness among concurrent video flows generated by heterogeneous client devices. For this purpose, a max-min fairness optimization problem is solved at runtime. We compare two approaches to actuate the optimal solution in an Software Defined Networking network: The first one allocates network bandwidth slices to video flows, and the second one guides video players in the video bitrate selection. We assess performance through several QoE-related metrics, such as Video Quality Fairness, video quality, and switching frequency. The impact of client-side adaptation algorithms is also investigated.

A hybrid model of adaptive video streaming control systems

Video streaming traffic over the Internet has significantly grown in the recent years. Adaptive video streaming control systems are employed to provide the best user experience given the user device and the network available bandwidth. The control goal is to maximize the video bitrate while avoiding playback interruptions. In this paper, we present a complete and accurate model of a generic adaptive streaming control system in the form of a hybrid dynamical system. The model describes all the system features, differently from previous models making the fluid-flow approximation, and allows to rigorously design video streaming controllers whose performance can be analytically assessed. The high accuracy of the model has been assessed by comparing numerical simulations to experimental data obtained through real network experiments. Given its accuracy and low computation cost, the proposed model provides a promising alternative to network experiments in order to aid the design and evaluation of adaptive video streaming systems.

Local SIP overload control: Controller design and optimization by Extremum Seeking

The session initiation protocol (SIP) is a signaling protocol for managing various types of real-time sessions between parties over an Internet Protocol (IP) network. An open issue is the control of overload situations that occur when the incoming flow of requests to a SIP server overcomes the server processing capacity. In particular, call establishment times increase due to overload, which triggers retransmissions and causes a further increase of the total incoming flow of requests. This paper proposes an overload control system for regulating both the queue length and the CPU load of the SIP server. The proposed control system is made of two proportional-integral controllers tuned by minimizing a proper cost function using the Extremum Seeking algorithm. A real working implementation of the SIP overload controller has been made in the open-source SIP server Kamailio. A performance evaluation and comparison of the proposed controller with the main proposals existing in the literature have been carried out. The results show that the proposed control system counteracts overload situations and provides a goodput close to the optimal while maintaining low call establishment delays and retransmission ratios.