Development of wireless ad hoc networks, virtualization, P2P, and DHT techniques for applications in streaming media, social networks, Telemedicine, and Precision Agriculture
The RSVP project aims at the development of advanced computing platforms for diferent classes of distributed applications in the new and important areas of scalable distribution of continuous media (video and audio streaming), participative social networks, Telemedicine, and Precision Agriculture. Such platforms will be based on new techniques and mechanisms such as virtual machines, distributed hash tables, the Peer-to-Peer distributed model, and wireless Ad hoc networks. Next, we summarize our research efforts.
In the area of mobile systems, we are concentrated on the development of an original approach to building ad hoc networks and running distributed applications. Specifically, we introduced the Active Prefixes (APs) and built a new interest-centric communication protocol based on APs, which together provide the necessary support to develop and run distributed applications in mobile ad hoc networks (MANETs) named Radnets in a cost-effective way. Radnets have been tested in different classes of applications including social networks, Telemedicine, Network security, vehicular networks, UAVs, and Evironmental control.
Energy-efficient computer systems are making increasing use of processors that have multiple core units, DVFS, and virtualization support. However, current system clocks have not been usually designed to cope with the capacity of such mechanisms to decelerate/accelerate the passage of time, which increases the time drifts in the system and produces adverse side effects. As an alternative to the system clock, we developed an original virtual clock, named RVEC, with the property that the time count is strictly increasing and precise. Furthermore, we used RVEC to build a High-Precision Global Clock (HPGC) solution which is free from resynchronization.
Distributed Hash Tables (DHTs) have been used in several applications, but most DHTs have opted to solve lookups with multiple hops, to minimize bandwidth costs while sacrificing lookup latency. We developed D1HT, which has a peer-to-peer and self-organizing architecture and is able to maximize lookup performance with reasonable maintenance traffic, and a Quarantine mechanism to reduce overheads caused by volatile peers. Our extensive set of results showed that D1HT is able to provide a very effective solution for a broad range of environments, from large-scale corporate datacenters to widely deployed Internet applications.
We have developed the global video environment (GloVE) technology for large scale distribution of video on demand to large audiences. GloVE uses a novel dynamic data structure to implement an innovative stream-reuse algorithm with several advantages over conventional caching algorithms. Moreover, we have used GloVE to build scalable vod platforms that support HTTP-based adaptive streaming technologies, real-time video transmission directly from mobile devices to the video server, video delivery networks (VDN), and vod systems for trainning and education.
Conference Publications
Technical Reports
Theses (in portuguese)
Dissertations (in portuguese)
Patents
Financial support: FINEP/MCT National Agency