Princeton EDGE Lab

In a community detection problem, we are given a graph and the goal is to identify the nodes in the graph that have strong ties to each other, or belong to the so called a “community”. In the context of social network analysis, the graph refers to the social network; a community refers to a group of people who interact closely with each other, such as researchers on the same topic or college students living in the same dorm...

Read more »

There exist many different modern approaches to resource allocation in computing clusters, ranging from the simplistic (FIFO, highest priority first, and earliest deadline first) to the more complex (fair scheduling, etc). However, many of these algorithms focus on maximizing optimization problems that have very basic job utilities that are either constant or binary with respect to a deadline time. We extend this problem to look at the nonconvex problem of scheduling jobs with deadlines, with time-dependent utility functions.

Read more »

A full duplex wireless device is one that can transmit and receive at the same time in the same frequency band, and typically it requires at least one Tx and one Rx antenna. The key challenge in realizing such a device lies in the self-interference (SI) generated by the Tx antenna at the Rx antenna. As an example, consider a Wi-Fi signal with a transmit power of 20 dBm. A Tx–Rx antenna separation of about 6–8 inches results in a path loss of about 40 dBm (depending on channel characteristics), resulting in a self-interference of at least -20 dBm...

Read more »

In recent years, content traffic has become increasingly dominated by video. The large file sizes coupled with exorbitant increases in demand are putting strain on even the highest-capacity fiber links in content delivery networks (CDNs). The natural question arises as to how such congestion can be alleviated.

We investigate a practical solution to this problem, which involves developing a satellite-based overlay for terrestrial CDNs. Satellite is ideal in this situation...

Read more »

Cloud computing is fast emerging as the backbone of most distributed services. Cloud-platforms such as Amazon's EC2, S3 or Microsoft's Azure provide scalable, fault tolerant computing and storage platforms over which many distributed solutions can be deployed. In this proposal, we focus on interactive services, a class of cloud-based services in which each user interacts with the cloud at very high frequencies through applications such as video conferencing, document collaboration or online gaming...

Read more »

People have been arguing about fairness for thousands of years, including academics working in economics, politics, philosophy, and engineering. Though fairness can consequently mean many different things, we focus on a specific aspect of fairness: defining the fairness of an allocation of resources to different people. We use a mathematical theory of fairness that is based on four fairness axioms, which yield a unique family of functions that quantify the fairness of a given resource allocation. These functions unify several previously proposed fairness measures.

Read more »

Heterogeneity of wireless network architectures is increasingly becoming an important part of current and next-generation wireless networks. Simultaneously, mobile devices are increasingly equipped with multiple radio access technologies (RATs) that can connect to and switch between different access networks. In such heterogeneous wireless environments, an important question that arises is "How should a user select the best access network to connect to at any given time?" To envision the difficulty of this task, imagine mobile users as cars on a busy multilane highway...

Read more »

Online service providers (OSPs) worry about the end-to-end performance their customers experience, since even small increases in user-perceived latency can have significant impact on revenues. To improve performance and reliability, service providers typically run services out of multiple geographically-distributed service locations--i.e., data centers, caches, availability zones--each typically peering with multiple Internet Service Providers (ISPs). In addition, service providers also need to consider operational costs: large services may serve petabytes of data daily, leading to significant bandwidth and electricity costs...

Read more »

DataMi is motivated by the growth in mobile (and wired) demand for data and ISPs' (Internet Service Providers) increasing inability to meet this demand. To make it worse, the heavy usage concentrates on several peak hours in a day, forcing ISPs to overprovision according to that. Even charging by monthly usage overages, as AT&T and Verizon have started doing in 2011, will not mitigate that. Yet this fact can be exploited to help solve the problem of excessive demand...

Read more »

Two emerging trends of Internet applications, video traffic becoming dominant and usage-based pricing plans becoming prevalent, are at odds with each other. On one hand, videos, especially on high-resolution devices (e.g., iPhone 5, iPad, Android tablets), consume much more data than other types of traffic...

Read more »

Twitter provides in real time a huge volume of data on people's opinions towards various issues. One natural question to ask is what we can achieve with this data. Another question is how reliable are existing Twitter-based methods to gauge people's opinions...

Read more »

Chiang receives Waterman award

DataMi receives IP Acceleration Fund

Chiang gives AURP, FCC, EdTech talks