Projects

COST IC1405: Reversible computation - extending horizons of computing

Reversible computation is an emerging paradigm that extends the standard forwards-only mode of computation with the ability to execute in reverse, so that computation can run backwards as naturally as it can go forwards. It aims to deliver novel computing devices and software, and to enhance traditional systems by equipping them with reversibility. The potential benefits include the design of revolutionary reversible logic gates and circuits - leading to low-power computing and innovative hardware for green ICT, and new conceptual frameworks, language abstractions and software tools for reliable and recovery oriented distributed systems.

Landauer's Principle, a theoretical explanation why a significant proportion of electrical power consumed by current forwards-only computers is lost in the form of heat, and why making computation reversible is necessary and beneficial, has only been shown empirically in 2012. Hence now is the right time to launch a COST Action on reversible computation. The Action will establish the first European (and the world first) network of excellence to coordinate research on reversible computation. Many fundamental challenges cannot be solved currently by partitioned and uncoordinated research, so a collaborative effort of European expertise with an industrial participation, as proposed by this Action, is the most logical and efficient way to proceed.

For more information click here.

TERAFLUX

In the TERAFLUX project we study the future of Teradevice systems. These systems will expose a large amount of parallelism (1000+ cores) that cannot be exploited efficiently by current applications and programming models. The aim of this project is to propose a complete solution that is able to harness the large-scale parallelism in an efficient way. The main objectives of the TERAFLUX project are the programming model, compiler analysis, and a scalable, reliable, architecture based mostly on commodity components. Data-flow principles are exploited at all levels as to overcome the current limitations. This is a large project with 10 EU partners: Università degli Studi di Siena (Italy), Barcelona Supercomputing Center (Spain), CAPS (France), Hewlett Packard Labs (Spain), INRIA (France), MICROSOFT R&D (Israel), THALES (France), University of Augsburg (Germany), University of Cyprus (Cyprus), University of Manchester (UK). Our main responsibility in the TERAFLUX project is the Architecture Work-Package. This work is done in collaboration with Professor Paraskevas Evripidou (UCY). From the CASPER group we have currently two students working for the project: Andreas Diavastos (PhD student) and Constantinos Christofi (PhD student). More information can be found in here

Intel SCC

We have been awarded an Intel SCC research processor for our research from Intel Corporation. This processor has 48 cores and we have been awarded it as to test the scalability of database query processing as well as the programmability using the TFlux model. Given that the processor has characteristics of future large-scale many-core processors, we will be testing our proposed techniques on it. This processor has been provided to us as a loan from Intel Corporation. We have currently various undergraduate and graduate students working on the SCC system.

HiPEAC NoE

European Network of Excellence on High Performance and Embedded Architecture and Compilation. More information can be found in here