High Performance Distributed Systems (HPDS)
A common platform for the group’s research is distributed and parallel systems, their architecture, design, implementation and behaviour. The research is organized into research areas and projects.
Description of research activities
The strength of the HPDS-group is in fundamental research on advanced computing systems and distributed systems, and in multidisciplinary research applying the system research to compute dependent application areas, i.e. weather forecasting and life sciences. Central issues studied include scalable concurrency, efficient memory access mechanisms, concurrent data structures and algorithms, programming systems, run-time systems, energy efficient computing, systems support for human-computer interfaces, collaborative systems and resource sharing, and data management and processing for data intensive bioinformatics and genomics applications.
The group experiments and publishes on topics including applying many-core graphics-processing units to boost the performance of computations and visualizations, prototyping multi-user, touch- and device-free human-computer interaction, large interactive high-resolution visualizations, tiled display walls, advanced computing systems, distributed systems, interactive weather forecasts, life sciences, interactive distributed performances, big data, and collaborative systems.
The display wall laboratory is a cornerstone of the HPDS-group. The display wall laboratory was initiated in 2004. It is comprised of a wall-sized, high-resolution tiled display with 28 projectors driven by a 28-node display cluster for a total of 22M-pixels.
The group has advanced the state of the art with its research on pixel distribution, compression, and caching to increase the visualization performance of typical standard desktop applications enabling them to visualize their output at an order of magnitude better resolution than what a single PC can support. The group has also developed several display-wall aware applications and systems to experimentally document how to do better with regards to both frame-rates and processing vs. existing architectures and designs for high performance visualizations.
The systems developed are used in various applications, including games, in an application supporting interactive viewing and navigation of a tiled 13Giga-pixel image over the City of Tromsø, and in a system transforming a standard 1.2M-pixel laptop into a 22M-pixel laptop.
The group's research is organized into research areas:
- Advanced computing systems
- High-performance visualization at high resolutions
- Collaborative systems and resource sharing - The group has developed scalable and flexible approaches and systems to do placeshifting and has documented their characteristics.
- Systems support for multi-user distributed human-computer interfaces - The group has researched and implemented an optical sensor based object detection and localisation system that accurately locates objects in 3D in front of a display wall. The system is used to do multi-user touch- and device-free interaction with applications using the display wall.
- Parallel, multi- and many-core systems - The group conducts research on load-time and run-time orchestration of applications, including instrumentation along the communication paths of a parallel multi-process and multi-threaded application.
- Cross-disciplinary Weather forecasting - To get useful level of detail in a weather forecast, the size of the forecasted geographical region is traded off against the available compute resources.
- Cross-disciplinary Life Sciences - The research is motivated by the extreme size of data sets produced by the life sciences. There is a need for better infrastructure systems for next-generation bioinformatics data exploration tools.
- Cross-disciplinary Opera - how to do systems support for a distributed opera with live performances taking place at several geographically distant locations
The group's webpage:
Videos of the display wall:
Other groups at the Department of Computer Science: