Y. Richard Yang

P4p: provider portal for applications

As peer-to-peer (P2P) emerges as a major paradigm for scalable network application design, it also exposes significant new challenges in achieving efficient and fair utilization of Internet network resources. Being largely network-oblivious, many P2P applications may lead to inefficient network resource usage and/or low application performance. In this paper, we propose a simple architecture called P4P to allow for more effective cooperative traffic control between applications and network providers. We conducted extensive simulations and real-life experiments on the Internet to demonstrate the feasibility and effectiveness of P4P. Our experiments demonstrated that P4P either improves or maintains the same level of application performance of native P2P applications, while, at the same time, it substantially reduces network provider cost compared with either native or latency-based localized P2P applications.

Maple: simplifying SDN programming using algorithmic policies

Software-Defined Networking offers the appeal of a simple, centralized programming model for managing complex networks. However, challenges in managing low-level details, such as setting up and maintaining correct and efficient forwarding tables on distributed switches, often compromise this conceptual simplicity. In this pa- per, we present Maple, a system that simplifies SDN programming by (1) allowing a programmer to use a standard programming language to design an arbitrary, centralized algorithm, which we call an algorithmic policy, to decide the behaviors of an entire network, and (2) providing an abstraction that the programmer-defined, centralized policy runs, conceptually, afresh on every packet entering a network, and hence is oblivious to the challenge of translating a high-level policy into sets of rules on distributed individual switches. To implement algorithmic policies efficiently, Maple includes not only a highly-efficient multicore scheduler that can scale efficiently to controllers with 40+ cores, but more importantly a novel tracing runtime optimizer that can automatically record reusable policy decisions, offload work to switches when possible, and keep switch flow tables up-to-date by dynamically tracing the dependency of policy decisions on packet contents as well as the environment (system state). Evaluations using real HP switches show that Maple optimizer reduces HTTP connection time by a factor of 100 at high load. During simulated benchmarking, Maple scheduler, when not running the optimizer, achieves a throughput of over 20 million new flow requests per second on a single machine, with 95-percentile latency under 10 ms.

Shadow configuration as a network management primitive

Configurations for todays IP networks are becoming increasingly complex. As a result, configuration management is becoming a major cost factor for network providers and configuration errors are becoming a major cause of network disruptions. In this paper, we present and evaluate the novel idea of shadow configurations. Shadow configurations allow configuration evaluation before deployment and thus can reduce potential network disruptions. We demonstrate using real implementation that shadow configurations can be implemented with low overhead.

A General Algorithm for Interference Alignment and Cancellation in Wireless Networks

Physical layer techniques have come a long way and can achieve very close to Shannon capacity for point-to-pint links. It is apparent that, to further improve network capacity significantly, we have to resort to concurrent transmissions. Many concurrent transmission techniques (e.g., zero forcing, interference alignment and distributed MIMO) are proposed in which multiple senders jointly encode signals to multiple receivers so that interference is aligned and each receiver is able to decode its desired information. In this paper, we investigate the constraints and challenges of using interference alignment. Our main contribution is conducting the first systematic investigation on the key issue of identifying opportunities for interference alignment. We identify diverse, novel scenarios for using interference alignment. We show that identifying opportunities for interference alignment in the general case is computational challenging. However, we also present a promising, distributed algorithm for identifying a wide range of opportunities for interference alignment using a unifying framework based on the degree of freedom. Our second contribution is evaluating key practical implementation issues.

Tango: Simplifying SDN Control with Automatic Switch Property Inference, Abstraction, and Optimization

A major benefit of software-defined networking (SDN) over traditional networking is simpler and easier control of network devices. The diversity of SDN switch implementation properties, which include both diverse switch hardware capabilities and diverse control-plane software behaviors, however, can make it difficult to understand and/or to control the switches in an SDN network. In this paper, we present Tango, a novel framework to explore the issues of understanding and optimization of SDN control, in the presence of switch diversity. The basic idea of Tango is novel, simple, and yet quite powerful. In particular, different from all previous SDN control systems, which either ignore switch diversity or depend on that switches can and will report diverse switch implementation properties, Tango introduces a novel, proactive probing engine that infers key switch capabilities and behaviors, according to a well-structured set of Tango patterns, where a Tango pattern consists of a sequence of standard OpenFlow commands and a corresponding data traffic pattern. Utilizing the inference results from Tango patterns and additional application API hints, Tango conducts automatic switch control optimization, despite diverse switch capabilities and behaviors. Evaluating Tango on both hardware switches and emulated software switches, we show that Tango can infer flow table sizes, which are key switch implementation properties, within less than 5% of actual values, despite diverse switch caching algorithms, using a probing algorithm that is asymptotically optimal in terms of probing overhead. We demonstrate cases where routing and scheduling optimizations based on Tango improves the rule installation time by up to 70% in our hardware switch testbed.

Guide to Reliable Internet Services and Applications

This guide addresses the challenges faced by service providers and the approaches they use to deliver reliable networks and networked application services. The book offers a systematic, inter-disciplinary approach and coverage of practical problems arising in real, operational deployments. Leading practitioners and researchers present their perspectives, and provide best practices. Features: Introduces the challenges of building reliable networks and services; Examines network reliability modeling and network planning; Investigates inter-domain reliability and overlay networks; Explores the critical function of network configuration management; Discusses network measurement and performance monitoring; Covers network management systems; Presents an approach to the design of reliable network application software, and an overview of server capacity and performance engineering. This text is suitable for an advanced undergraduate or graduate course, and will be of value to researchers and practitioners.

THash: A Practical Network Optimization Scheme for DHT-based P2P Applications

P2P platforms have been criticized because of the heavy strain that they can inflict on costly inter-domain links of network operators. It is therefore mandatory to develop network optimization schemes for controlling the load generated by a P2P platform on an operator network. While many research efforts exist on centralized tracker-based systems, in recent years multiple DHT-based P2P platforms have been widely deployed and considered as commercial services due to their scalability and fault tolerance. Finding network optimization for DHT-based P2P applications has thereby potential large practical impacts. In this paper, we present THash, a simple scheme that implements a distributed and effective network optimization for DHT systems. THash uses standard DHT put/get semantics and utilizes a triple hash method to guide the DHT clients to choose their sharing peers in proper domains. We have implemented THash in a major commercial P2P system (PPLive), using the standard ALTO/P4P protocol as the network information source. We conducted experiments over this network in real operation and observed that compared with Native DHT, THash reduced respectively by 47.4% and 67.7% the inter-PID and inter-AS traffic, while reducing the average downloading time by 14.6% to 24.5%.

FAST: A Simple Programming Abstraction for Complex State-Dependent SDN Programming

Handling state dependencies is a major challenge in modern SDN programming, but existing frameworks do not provide sufficient abstractions nor tools to address this challenge. In this paper, we propose a novel, high-level programming abstraction and implement the *Function Automation SysTem (FAST)*. With the two key features, i.e., *automated state dependency tracking* and *efficient re-execution scheduling*, we demonstrate that FAST substantially simplifies state-dependent SDN programming and boosts the performance.

The Challenges of Building Reliable Networks and Networked Application Services

In the decades since the ARPANET interconnected four research labs in 1969 [1], computer networks have become a critical infrastructure supporting our information-based society. Our dependence on this infrastructure is similar to our dependence on other basic infrastructures such as the world’s power grids and the global transportation systems. Failures of the network infrastructure or major applications running on top of it can have an enormous financial and social cost with serious consequences to the organizations and consumers that depend on these services.

Design of a Distributed Flow Control Scheme Based on Wireless Multi-Rate Multicast Networks

With the ever-increasing wireless multicast data applications recently, considerable efforts have focused on the design of self-adaptive flow control schemes for wireless multicast service. This paper proposes a novel and efficient distributed flow control scheme for wireless multi-rate multicast (MR-M), based on the well-known Proportional Integral and Derivative (PID) controllers. The PID controller at each router computes its expected incoming rate and feedbacks this rate to its upstream router, such that the local buffer occupancy can be stabilized at an appropriate value. We give the theoretical analysis of the proposed PID controller in terms of system stability. The proposed MR-M controller achieves the fairness in two aspects: 1) The intra-session fairness, i.e., the receivers from the same source within the same multicast session, if they subscribe networks with different capacities, can receive data at different rates; 2) The inter-session fairness, i.e., the link bandwidth is fairly shared among multiple multicast sessions from different sources. Extensive simulations have been conducted and the results have demonstrated a superior performance of the proposed scheme based on system stability, high link utilization, high throughput.