Despite increasing mobile Internet penetration in emerging regions, the growth of web page complexity combined with inadequate content delivery mechanisms and lack of relevant local content make the web experience poor for users in these regions. In this paper, we propose GAIUS, a content ecosystem that enables the efficient creation and dissemination of locally relevant web content. GAIUS aims at creating a community content exchange that enables users to easily create and consume relevant web content with the help of a mobile application. Based on an early deployment of GAIUS in various regions, we show that our ecosystem enables localized content creation and diffusion efficiently.

Rural areas are home to 45% of the world's population and are neglected in the design and deployment of broadband Internet access. Beyond establishing connectivity, the planning and design of networks in these areas is often challenging because the complexity of planning such networks often exceeds the knowledge and practical expertise of individuals -- even individuals that are experts in this field. Yet, proper planning and efficient network designs are critical to the long term viability and scalability of such networks. In this paper, we highlight the challenges that lone operators who lack sufficient technical support and tools must face when designing and deploying their networks. We then present a network planning tool, Zyxt, that aims to help individuals in planning rural wireless networks. We describe the design and implementation of our prototype and evaluate it against synthetic and real-world planning problems. Our evaluation demonstrates that Zyxt is able to solve real-world network planning problems using relatively modest computing resources within a few hours.

End-users in emerging markets experience poor web performance due to a combination of three factors: high server response time, limited edge bandwidth and the complexity of web pages. The absence of cloud infrastructure in developing regions and the limited bandwidth experienced by edge nodes constrain the effectiveness of conventional caching solutions for these contexts. This paper describes the design, implementation and deployment of xCache, a cloud-managed Internet caching architecture that aims to proactively profile popular web pages and maintain the liveness of popular content at software defined edge caches to enhance the cache hit rate with minimal bandwidth overhead. xCache uses a Cloud Controller that continuously analyzes active cloud-managed web pages and derives an object-group representation of web pages based on the objects of a page. Using this object-group representation, xCache computes a bandwidth-aware utility measure to derive the most valuable configuration for each edge cache. Our preliminary real-world deployment across university campuses in three developing regions demonstrates its potential compared to conventional caching by improving cache hit rates by about 15%. Our evaluations of xCache have also shown that it can be applied in conjunction with other web optimizations solutions like Shandian, and can improve page load times by more than 50%.

While the networking literature is replete with work on managing and operating networks---from the specifics of protocols to the design of management tools and architectures---there is comparatively little work on planning a network to be rolled out. In part this is because the task of network planning typically falls to carriers (for backbones) and cloud providers (for datacenters), which have the resources and the control to meet their specific needs. Here we consider network planning in situations that are quite different: resource poor and highly constrained.

Modern web pages are very complex; each web page consists of hundreds of objects that are linked from various servers all over the world. While mechanisms such as caching reduce the overall number of end-to-end requests saving bandwidth and loading time, there is still a large portion of content that is re-fetched -- despite not having changed. In this demo, we present Extreme Cache, a web caching architecture that enhances the web browsing experience through a smart pre-fetching engine. Our extreme cache tries to predict the rate of change of web page objects to bring cacheable content closer to the user.

Legacy congestion controls including TCP and its variants are known to perform poorly over cellular networks due to highly variable capacities over short time scales, self-inflicted packet delays, and packet losses unrelated to congestion. To cope with these challenges, we present Verus, an end-to-end congestion control protocol that uses delay measurements to react quickly to the capacity changes in cellular networks without explicitly attempting to predict the cellular channel dynamics. The key idea of Verus is to continuously learn a delay profile that captures the relationship between end-to-end packet delay and outstanding window size over short epochs and uses this relationship to increment or decrement the window size based on the observed short-term packet delay variations. While the delay-based control is primarily for congestion avoidance, Verus uses standard TCP features including multiplicative decrease upon packet loss and slow start. Through a combination of simulations, empirical evaluations using cellular network traces, and real-world evaluations against standard TCP flavors and state of the art protocols like Sprout, we show that Verus outperforms these protocols in cellular channels. In comparison to TCP Cubic, Verus achieves an order of magnitude (> 10x) reduction in delay over 3G and LTE networks while achieving comparable throughput (sometimes marginally higher). In comparison to Sprout, Verus achieves up to 30% higher throughput in rapidly changing cellular networks.

Web access is prohibitively slow in many developing regions despite substantial effort to increase bandwidth and network penetration. In this paper, we explore the fundamental bottlenecks that cause poor web performance from a client's perspective by carefully dissecting webpage load latency contributors in Ghana. Based on our measurements from 2012 to 2014, we find several interesting issues that arise due to the increasing complexity of web pages and number of server redirections required to completely render the assets of a page. We observe that, rather than bandwidth, the primary bottleneck of web performance in Ghana is the lack of good DNS servers and caching infrastructure. The main bottlenecks are: (a) Recursive DNS query resolutions; (b) HTTP redirections; (c) TLS/SSL handshakes. We experiment with a range of well-known end-to-end latency optimizations and find that simple DNS caching, redirection caching, and the use of SPDY can all yield substantial improvements to user-perceived latency.