| Internet streaming services are very popular today. According to Cisco, video traffic accounts for more than 51% total Internet traffic. With the pervasive adoption of various mobile devices in practice in the past several years, today Internet streaming services are receiving a rapidly growing number of requests from various mobile devices. As a result, more than 50% of the data consumed by mobile devices is video streaming traffic. However, streaming delivery to mobile devices is more challenging than to its desktop counterpart.;In this dissertation, we first empirically investigate Internet mobile streaming practices. For this purpose, we conduct detailed analysis on a one-month server log (with 212 TB delivered video traffic) from a top Internet mobile streaming service provider serving world-wide mobile users. We investigate mobile streaming from various perspectives, including hardware and software heterogeneity, different characteristics of mobile videos, and different user access patterns. The results provide us in-depth understanding on the current Internet mobile streaming services.;A critical constraint on mobile devices for receiving Internet streaming services is that they have limited battery capacities. While watching streaming videos, the battery power is depleted at a fast rate by the wireless network interface card (WNIC) for streaming data transmission, by the CPU or GPU for computationally intensive video decoding, and by the display for rendering the video. Among these power-consuming sources, transmission power consumption is very significant: for a mobile device receiving streaming services, about 30% to 40% of the power is consumed by the WNIC for streaming data transmission. So in order to prolong the battery lifetime, it is important to save the battery power consumed by the WNIC. For this purpose, we design and implement new schemes that can effectively save battery power consumption while maintaining good streaming quality to mobile devices. In particular, we focus on peer-to-peer (P2P) streaming and client-server streaming as they are widely used. We aim to save battery power consumption from two aspects: (1) how the data is received, and (2) how much data is received.;Through extensive Internet experiments, we find that the widely used 802.11 power saving mode (PSM) cannot effectively save power for popular P2P streaming services: several unique characteristics of P2P traffic, such as highly-frequent and delay-sensitive control packets, often prevent the wireless interface from switching into power saving mode. To overcome these challenges, we design BlueStreaming, a system that can intelligently leverage Bluetooth to transmit P2P control traffic. Experiments using a BlueStreaming prototype with a number of popular P2P streaming applications show that BlueStreaming can achieve a reduction of power consumption by up to 46% compared to the commodity PSM scheme.;Besides how data is delivered, the battery power consumption on mobile devices is also affected by how much data is received. However, in the current practice, we find from experiments that iOS users always receive a significant amount of redundant traffic when accessing Internet streaming services such as YouTube in a client-server architecture. Such redundant traffic not only leads to more battery power consumption but also can incur monetary cost. After analyzing the underlying reasons, we design and implement a system called CStreamer that can minimize such redundant traffic. In experiments, a CStreamer prototype deployed on Amazon EC2 completely eliminates redundant traffic and reduces battery power consumption by up to 40%. |
