Projects / Programmes
Parallel and Distributed Systems
January 1, 2014
- December 31, 2016
| Code |
Science |
Field |
Subfield |
| 2.07.00 |
Engineering sciences and technologies |
Computer science and informatics |
|
| 2.08.00 |
Engineering sciences and technologies |
Telecommunications |
|
| Code |
Science |
Field |
| P170 |
Natural sciences and mathematics |
Computer science, numerical analysis, systems, control |
| Code |
Science |
Field |
| 1.02 |
Natural Sciences |
Computer and information sciences |
| 2.02 |
Engineering and Technology |
Electrical engineering, Electronic engineering, Information engineering |
Researchers (23)
Organisations (3)
Abstract
Objectives: We need more powerful computers in all areas of human activity: science, engineering, medicine, economics, finances, and industry. We believe that the future challenges are in the parallel and distributed computing systems. Parallel computers have already penetrated into the field of mobile devices and will resolve most of the algorithmic tasks in the future. They are also gaining momentum in the fields of artificial intelligence and expert systems. Knowledge is the component that converts the parallel computer into a useful device, so the field is beneficial also for the industry.
Goals: We expect to utilize the powerful computers for compute intensive simulations of natural phenomena, information and technology processes, which cannot be fully explored by experiments only. We expect to solve compute intensive practical problems more efficiently, using new parallel and distributed procedures, and modern methods (randomization, approximate nature, mimicking the natural phenomena). Our goal is also effective management and analysis of large data streams.
Expected results: The results of our research programme group will contribute to more efficient use of computing resources, shorter execution time of compute intensive tasks, and better understanding and interpretation of the theoretical foundations of distributed and parallel computing. We will discover new knowledge and learn new skills for the implementation of efficient parallel and distributed computers. We will strengthen our role in the integration of research within the study programs of Slovenian and foreign universities. We will help increase the competitiveness of our industry, which should be based on an engineering development and the added knowledge. On the other hand, our newly developed methods will contribute to improved human health and living environment.
Significance for science
The importance of computers is undeniable. They form the standard equipment in most areas of technical research and are standard tools for complex tasks, such as system management, global financial market, medical diagnostics, exploration of the Earth and the universe, and many more. The complexity of the tasks that are being solved on computers has been growing rapidly, as demands for increasingly precise and accurate numerical predictions are becoming higher and higher. As a result, computer algorithms are becoming more expensive, both from the financial and the temporal aspect and, last but not least, the ecological point of view. Often, a real-time response also plays a crucial role, which typically results in either additional expenses or even prevents the practical computer applications altogether. Obtaining expensive high-performance computer systems is often recognized as a remedy for the increasing demands. However, the raw computational power is in most cases poorly exploited, especially since it comes on the account of parallel architectures, whether these are multi-processors, multi-cores, processor accelerators or computing clusters. The technological and physical limitations do not allow for any other means of Increasing the computational power. Currently only massively parallel computers with more than 10^15 Floating point Operations Per Second (PFLOPS), i.e. Petascale, allow achieving Moore's law expectations (doubling the computing performance in each two years). Assuming that one floating point operation requires one system clock period, a sequential computer with a 10 GHz system clock would reach "just" 10^10 FLOPS, which is far below the Petascale. Today, the only option to break this limit in approaching to Exascale era (10^18 FLOPS), is a parallel computer with millions of processors that run in an optimal, parallel way. However, parallel execution of a program is limited by the speed of the communications between the processors and the ability of the algorithms to be parallelized. Every part of the code that cannot be written in parallel becomes a bottleneck in the parallel efficiency. This gravely limits the scalability of many problems that should be parallelized. On the other hand, the excessive power consumption, which is today at about 0.5 kW/TFLOPS, represents an environmental and financial problem, and is becoming an increasingly important factor in information technologies. Therefore, it is an imperative to find procedures and methods for even more efficient utilization of cooperating computing resources. A deeper knowledge is needed in computer architectures, procedures for communication protocols, data reconciliation, access to shared resources, error detection, and similar themes, which are central to our programme. Based on such knowledge, we can achieve new breakthroughs in the field of computer science, in order to significantly shorten calculation times and to reduce the energy consumption. The proposed research programme aims at theoretical investigations in the field of parallel and distributed processing. The research will be focused on parallel algorithms, on methods and frameworks for their development, better methods for programming, modelling, standardization, and improved exploitation of distributed systems and communication networks. New findings are expected particularly in the development of new parallel algorithms and simulation models, in temporal and spatial complexity analysis and in the management of processing and communication resources. Research on protocols for the use of common variables and cooperation on common tasks will contribute to the integration and standardization and thereby have important role in future information services. Theoretical solutions and procedures will be conceived in a way that will facilitate their wide application.
Significance for the country
Immediate importance of distributed systems research is in the fact that it represents a basis for the development of Slovenia, as it facilitates for various applications better access to large data sets and network resources. Distributed systems also increase the benefit of the past investments into the network infrastructure. Computationally intensive problems become more manageable in practice through the better understanding of theoretical foundations. Thereby one can solve larger and more difficult problems, which are closer to models of real-world processes. Slovenia has built advanced information-communication infrastructure, through which it has good chances of becomeing an important player in the field of ICT. The applicability of our past research in the field is evident from the overview of the main trends in the development of computing and ICT. Internet is becoming the most important medium, also in industrial area, and computer industry is becoming one of the most important industrial branches. We have recognized already before these trends in which direction the development will extend. The programme members helped in establishing of the first Slovenian inter-institutional test GRID and developed an environment and example applications immediately useful for smaller enterprises, which could thereby improve the exploitation of their ICT resources. Within the programme, various devices and modules were developed for the integration of data sources, sensor networks and video information processing. Contributions to efficient network computing and to economic energy consumption were made and the implementation of projects with high added value, which often have to be supported by HPC simulations, was made possible. In collaboration with medical field, new methods for more efficient treatment and prevention of dangerous diseases, particularly in the field of neurology, cardiology and general practice were developed and tested. Devices for the measurement of bio-signals, indispensable for the development of new diagnostic methods were prototyped. The acquired knowledge is applied, for example, in the area of e-medicine and efficient power consumption. Beside the mentioned impact, our research also has indirect importance for the Slovenian development. Research results are promptly presented at workshops, scientific conferences and exhibitions. The main results are published in the principal journals of our research field. This represents Slovenia as a country which is significantly contributing to the knowledge of the mankind. We participate also in international projects and in the organization of international conferences and summer schools. We see significant indirect importance also in the introduction of new study areas at the University of Ljubljana, from which a part of our programme group originates. Research work in our group contributes to better quality of study programs in the field. Parallel and distributed computing, on which our research is focused, is a long-term technological need of the ICT sphere of governmental institutions and industry. Through new hypotheses, new problems requiring research are constantly identified. To competently participate in research ad development, it is crucial to follow the fast development of ICT. The development of computer and network technologies stimulates the technological and professional development also in other fields. The programme contributes to achieving new understanding, implementing improvements in medical diagnostics a higher competitiveness of the industry.
Most important scientific results
Annual report
2014,
2015,
final report
Most important socioeconomically and culturally relevant results
Annual report
2014,
2015,
final report
