Department of Computer Science

Distributed and Pervasive Systems Research Focus

We envision the accelerated convergence of mobile and cloud computing technologies, and the maturation of big data in the coming years.

Compute-intensive technologies like natural language processing, computer vision, 3D computer graphics, and machine learning will be widely applied in the emerging mobile applications. Furthermore, the focus of mobile sensing will shift from individual sensing towards crowdsensing, and the usage of data-intensive sensors such as cameras will increase rapidly, along with the pervasive availability of intelligent wearable devices such as Google Glasses. Accordingly, we are extending our current research on energy-efficient mobile computing, distributed cloud computing, and the Internet of Things, to solve the challenges posed by the emerging mobile systems, applications, and services.

Energy-aware computing and communications are a very timely and important topic. Energy consumption is a concern with today’s mobile devices. While the capabilities of the devices have improved rapidly over the last ten years transforming also the way these devices are being utilized, battery technology has not been able to keep up with this evolution. As a consequence, there is an increasing gap between the battery capacity and the amount of energy required for typical usage. In addition, reducing energy consumption and carbon footprint has been widely recognized as a challenge for the whole ICT industry. The growing operational expenditure in ICT calls for more energy-aware networking solutions throughout the entire end-to-end communication chain. We address these challenges by building models of energy consumption through experiments and measurements based on which we develop new more energy-efficient protocols and services.

Distributed systems and services address architectures, platforms, and protocols for flexible, scalable, and easily usable services. Cloud computing uses virtual resources in the Internet for computing and storage, and is able to elastically scale to match changing resource needs. Merging cloud technologies with mobile domain has potential to offer new technology innovations and business opportunities for operators, vendors, and developers, as well as novel services for end users. Mobile cloud computing provides versatile research topics varying from the virtualization layer up to the service layer. Machine-to-machine communication, wireless sensor networks, and peer-to-peer applications are other examples of active research areas in the distributed services domain.

We expect two trends to be increasingly dominant in the future. First, data storage and processing will increasingly happen in massive-scale distributed systems through cloud computing. Second, more and more services are used on mobile devices, which besides mobile phones and tablets increasingly mean pervasive or wearable devices. We are already witnessing the Internet of Things entering the everyday life of ordinary people through pervasive devices such as activity tracking wristbands, smart watches, Google Glasses, cars with Internet connectivity, and so on. As a consequence, mobile communication systems will need to scale up to tens of billions of nodes when these new devices are connected to applications and services provided by mobile communication systems and the cloud computing infrastructures behind them.

The Distributed and Pervasive System area has the mission to address such challenges. Research is targeting the development of a distributed cloud infrastructure to provide high performance and end-to-end optimized solutions for latency- and storage-sensitive, as well as bandwidth-intensive, storage-sensitive, or and compute-intensive mobile cloud applications. Specifically, the major goal is to provide modular re-usable instruments for distributed mobile computing systems, and utilizing such infrastructure to solve the challenges to mobile crowdsensing and other emerging applications. To this end, new methodologies and techniques will be developed for scalable mobile crowdsensing and intelligent data analysis of large-scale heterogeneous datasets in ubiquitous cloud environment. The use cases will include vertical applications and services like crowdsourced video and imaging, smart traffic, efficient storage and processing of Big Data, including novel distributed data processing frameworks.

  • Published:
  • Updated: