Navigation

News

Quick Links

Scientific Conferences and Workshops

Here is a list of conferences of interest to researchers in the areas of compilers, parallel processing, and supercomputing. It is maintained primarily for the benefit of the Computing Systems Lab at the National Technical University of Athens, though it will hopefully be of use to others.

Research Groups at Computing Systems Laboratory

Parallel and Distributed Computing Systems

The main objective of this activity is to combine the advantages of distributed memory architectures (scalability) and the advantages of the shared memory programming paradigm (easy and secure programming). Other relevant activities are the automatic extraction of parallelism and the subsequent mapping of algorithms onto different types of parallel architectures.

In this context, our research has focused on applying transformations to nested for-loops, in order to efficiently execute them in Non-Uniform Memory Access (NUMA) machines, such as the SCI-Clusters of the laboratory. In particular, we apply a transformation called tiling, or supernode transformation in order to minimize the communication latency effect on the total parallel execution time of the algorithms. Tiling method groups neighboring computation points of the nested loop into blocks called tiles or supernodes thus increasing the computation grain and decreasing both the communication volume and frequence. Applying the tiling techniques, we have developped a tool, which accepts C-like nested loops and partitions them into groups/tiles with small inter-communication requirements. The tool automatically generates efficient message passing code (using MPI) to be executed on SMPs or clusters. Future work contains comprarisons of certain variations of tiling (shape, size, etc) and code generation tecniques based on experimental results taken from the application of the tool on an SCI-cluster.

In addition, we explore several methods such as overlapping of communication and computations, in order to further reduce the total execution time of the transformed code. The targeted communication platforms include SCI, GM message passing over Myrinet interconnect, etc.

High performance web servers using linux clusters

Currently the group is performing research on high performance web serving using clusters of PCs and more specifically the linux SCI cluster the group has assembled. Our research focuses on the implementation of a distributed filesystem over the DSM SCI provides. This memory based filesystem will be used by the web server for fast file retrieval, resulting to an overall speed up (reduction of disk I/O).

High Level Hardware Design Automation

The automatic transformation of high-level algorithmic specifications of computing problems into a specific architectural configuration is the major purpose of this activity. New methodologies are investigated, using Prolog and FP formalisms for the behavioral specification of the algorithms to be mapped onto application-specific architectures.

Neural Network Theory and Parallel Processing

The interests of this research group include areas of parallel and distributed computing, neural network algorithms and architectures, fault tolerance, intelligent planning and optimization techniques, performance modelling, as well as topics related to image processing and pattern recognition, learning techniques, storage and retrieval of pictorial information. The Group is active in R and D programmes sponsored by Greek and European organisations and possesses considerable experience in the above fields, both in research and applications.

Medical Applications

The purpose of this activity is to develop integrated medical applications, regarding the manipulation of electrocardiogram (ECG) signals. It refers to the recognition, measurement, diagnosis and transmission of ECG signals. Moreover, we are engaged in the development of specialised training software for ECG interpretation, using multimedia technology.

Multimedia Systems in CAI, CAT

This activity aims to the exploration of modern multimedia technology in the development of interactive courseware, for applications in basic education and in vocational training. The laboratory has acquired several workstations enhanced by multimedia equipment and appropriate authoring tools. It is also augmented by a CD-R formatting facility.

Research Projects

SMART-PIV

"Development of an Interactive Integrated P.I.V System based on Miniaturised Optical Sensor Technology for Implantable Biomedical Devices Design," IST-2002-37548 (to commence in June 2002, role of cslab is to provide HPC middleware to the project).

TOPCARE

"Implementation of a telematic homecare platform in cooperative health care provider networks," IST-2000-25068.

INCO-DC

"PEACE by High Performance Computing," INCO-950895-EU.

TetraMED

"Technology Transfer in HPCN in the Mediterranean Countries," HPCN-III/98/028-EU.

VELOS

"Advanced CAD Applications in Aircraft and Vehicle Design using High Performance Computing Architectures," GSRT-EKBAN.

PENED-91/663

"Automated Synthesis of VLSI Architectures for Special Purpose Aplications", Greek Secretariat for Research and Technology-GSRT, Ministry of Development.

PENED-95/1405

"Automatic Parallelization of Sequential Algorithms", Greek Secretariat for Research and Technology"-GSRT, Ministry of Development.

AMBULANCE

"Mobile Unit for Health Care Provision via Telematics Support." (EU/Telematics - Health Care).

NIKA

"A generalised system for medical image management," GSRT-EKBAN-504.

HILDE

"Hypermedia Intelligent Learning Design Environment," GSRT--2.

MODULATES MM-1018

"Multimedia Organisation for Developing the Understanding and Learning of Advanced Technology in European Schools," EU-Educational Multimedia Taskforce.

ARTEMIS

"Greek Distributed Digital Library for Grey Literature," EPEAEK.

DIALOGOS

"Development of the Greek language and speech industry," GSRT-EKBAN (parallel computing platform for Greek speech recognition).