Te-Yuan Huang

Carving research slices out of your production networks with OpenFlow

1. SLICED PROGRAMMABLE NETWORKS OpenFlow [4] has been demonstrated as a way for researchers to run networking experiments in their production network. Last year, we demonstrated how an OpenFlow controller running on NOX [3] could move VMs seamlessly around an OpenFlow network [1]. While OpenFlow has potential [2] to open control of the network, only one researcher can innovate on the network at a time. What is required is a way to divide, or slice, network resources so that researchers and network administrators can use them in parallel. Network slicing implies that actions in one slice do not negatively affect other slices, even if they share the same underlying physical hardware. A common network slicing technique is VLANs. With VLANs, the administrator partitions the network by switch port and all traffic is mapped to a VLAN by input port or explicit tag. This coarse-grained type of network slicing complicates more interesting experiments such as IP mobility or wireless handover. Here, we demonstrate FlowVisor, a special purpose OpenFlow controller that allows multiple researchers to run experiments safely and independently on the same production OpenFlow network. To motivate FlowVisor’s flexibility, we demonstrate four network slices running in parallel: one slice for the production network and three slices running experimental code (Figure 1). Our demonstration runs on real network hardware deployed on our production network at Stanford and a wide-area test-bed with a mix of wired and wireless technologies.

A buffer-based approach to rate adaptation: evidence from a large video streaming service

Existing ABR algorithms face a significant challenge in estimating future capacity: capacity can vary widely over time, a phenomenon commonly observed in commercial services. In this work, we suggest an alternative approach: rather than presuming that capacity estimation is required, it is perhaps better to begin by using only the buffer, and then ask when capacity estimation is needed. We test the viability of this approach through a series of experiments spanning millions of real users in a commercial service. We start with a simple design which directly chooses the video rate based on the current buffer occupancy. Our own investigation reveals that capacity estimation is unnecessary in steady state; however using simple capacity estimation (based on immediate past throughput) is important during the startup phase, when the buffer itself is growing from empty. This approach allows us to reduce the rebuffer rate by 10-20% compared to Netflixs then-default ABR algorithm, while delivering a similar average video rate, and a higher video rate in steady state.

Downton abbey without the hiccups: buffer-based rate adaptation for HTTP video streaming

Recent work has shown how hard it is to pick a video streaming rate. Video service providers use heuristics to estimate the network capacity leading to unnecessary rebuffering events and suboptimal video quality. This paper argues that we should do away with estimating network capacity, and instead directly observe and control the playback buffer. We present a class of rate selection algorithms that allow us to optimize the delivered video quality while provably never unnecessarily rebuffering. Our algorithms work with discrete video rates, video chunking and for both CBR and VBR video codecs.

Making use of all the networks around us: a case study in android

Poor connectivity is common when we use wireless networks on the go. A natural way to tackle the problem is to take advantage of the multiple network interfaces on our mobile devices, and use all the networks around us. Using multiple networks at a time makes makes possible faster connections, seamless connectivity and potentially lower usage charges. The goal of this paper is to explore how to make use of all the networks with todays technology. Specifically, we prototyped a solution on an Android phone. Using our prototype, we demonstrate the benefits (and difficulties) of using multiple networks at the same time.

Could Skype be more satisfying? a QoE-centric study of the FEC mechanism in an internet-scale VoIP system

The phenomenal growth of Skype in recent years has surpassed all expectations. Much of the applications success is attributed to its FEC mechanism, which adds redundancy to voice streams to sustain audio quality under network impairments. Adopting the quality of experience approach (i.e., measuring the mean opinion scores), we examine how much redundancy Skype adds to its voice streams and systematically explore the optimal level of redundancy for different network and codec settings. This study reveals that Skypes FEC mechanism, not so surprisingly, falls in the ballpark, but there is surprisingly a significant margin for improvement to ensure consistent user satisfaction.

A measurement study of zigbee-based indoor localization systems under RF interference

With an expected market value of 2.71 billion in 2016, supporting daily use of real-time location systems in households and commercial buildings is an increasingly important subject of study. A growing problem in providing robust indoor location estimations in real time is the use of wireless transmissions in RF frequencies. Having implemented a simple RSSI-signature-based location system on a 24-node IEEE 802.15.4-based sensor network testbed, we are able to analyze the effect of background IEEE 802.11 traffic on localization error. The measurement results demonstrate that the 80th-percentile of the localization error may increase by 141% at worst in an office building with active use of IEEE 802.11 for data. Such performance degradation results from RSSI reading loss as the beacon messages collide with background traffic.

Putting home users in charge of their network

Policy-makers, ISPs and content providers are locked in a debate about who can control the Internet traffic that flows into our homes. In this paper we argue that the user, not the ISP or the content provider, should decide how traffic is prioritized to and from the home. Home users know most about their preferences, and if they can express them well to the ISP, then both the ISP and user are better off. To test the idea we built a prototype that lets users express highlevel preferences that are translated to low-level semantics and used to control the network.

Tuning Skype's Redundancy Control Algorithm for User Satisfaction

Determining how to transport delay-sensitive voice data has long been a problem in multimedia networking. The difficulty arises because voice and best-effort data are different by nature. It would not be fair to give priority to voice traffic and starve its best-effort counterpart; however, the voice data delivered might not be perceptible if each voice call is limited to the rate of an average TCP flow. To address the problem, we approach it from a user-centric perspective by tuning the voice data rate based on user satisfaction. Our contribution in this work is threefold. First, we investigate how Skype, the largest and fastest growing VoIP service on the Internet, adapts its voice data rate (i.e., the redundancy ratio) to network conditions. Second, by exploiting implementations of public domain codecs, we discover that Skypes mechanism is not really geared to user satisfaction. Third, based on a set of systematic experiments that quantify user satisfaction under different levels of packet loss and burstiness, we derive a concise model that allows user-centric redundancy control. The model can be easily incorporated into general VoIP services (not only Skype) to ensure consistent user satisfaction. Index Terms—MOS, PESQ, Piggyback, QoE (Quality of Ex- perience), QoS (Quality of Service), VoIP

Scheduling packets over multiple interfaces while respecting user preferences

Now that our smartphones have multiple interfaces (WiFi, 3G, 4G, etc.), we have preferences for which interfaces an application may use. We may prefer to stream video over WiFi because it is fast, but VoIP over 3G because it gives continued connectivity. We also have relative preferences, such as giving Netflix twice as much capacity as Dropbox. This means our mobile devices need to schedule packets in keeping with our preferences while making use of all the capacity available. This is the natural domain of fair queuing, and this paper is about the design of a packet scheduler to meet these requirements. We show that traditional fair queueing schedulers cannot take into account a users preferences for some interfaces over others. We present a novel packet scheduler called miDRR that meets our needs by generalizing DRR for multiple interfaces. We demonstrate a prototype running in Linux and show that it works correctly and can easily run at the speeds we need.

PhoneNet: a phone-to-phone network for group communication within an administrative domain

This paper proposes PhoneNet, an application framework to support direct group communication among phones without relay nodes. PhoneNet presents the familiar abstraction of a multi-user chat service to application writers. It performs two main functions: inviting participants to the chat room and routing data between participants directly without going through any intermediaries. Made possible by a generic chat room service embedded in the network itself, all application-specific code in PhoneNet applications runs on the phones themselves. Unlike the conventional server-client model, this design does not require scalable central servers that can handle all simultaneous interactions. As a first step, we have created a prototype of PhoneNet that works within an administrative domain. The multicast functionality among phones is implemented on top of a software-defined network (SDN). We have developed two applications using PhoneNet: teleconferencing and photo-sharing. Our experience suggests that it is easy to develop PhoneNet applications and PhoneNet appears to be effective in reducing network traffic.