EXA2PRO : Enhancing Programmability and boosting Performance Portability for Exascale Computing Systems
Type of action: RIA
Call identifier: H2020-FETHPC-2017
The Software Engineering Group, of the Software and Data Engineering Laboratory, of the Department of Applied Informatics at the University of Macedonia, proudly announces its participation in the EXA2PRO project (as part of the CERTH group).
The vision of EXA2PRO is to develop a programming environment that will enable the productive deployment of highly parallel applications in exascale computing systems. EXA2PRO programming environment will integrate tools that will address significant exascale challenges. It will support a wide range of scientific applications, provide tools for improving source code quality, enable efficient exploitation of exascale systems’
heterogeneity and integrate tools for data and memory management optimization. Additionally, it will provide various fault-tolerance mechanisms, both user-exposed and at runtime system level and performance monitoring features. EXA2PRO will be evaluated using 4 use cases from 4 different domains, which will be deployed in JUELICH supercomputing center. The use cases will leverage the EXA2PRO tool-chain and we expect: – Increased applications performance based on EXA2PRO optimization tools (data and memory management) – Efficient exploitation of heterogeneity by the applications that will allow the evaluation of more complex problems. – Identification of trade-offs between design qualities (source code maintainability/reusability) and run-time constraints (performance/energy consumption). – Evaluation of various fault-tolerance mechanisms for applications with different characteristics. EXA2PRO outcome is expected to have major impact in a) the scientific and industrial
community that focuses on application deployment in supercomputing centers: EXA2PRO environment will allow efficient application deployment
with reduced effort. b) on application developers of exascale application: EXA2PRO will provide tools for improving source code maintainability/reusability, which will allow application evaluation with reduced developers’ effort. c) on the scientific community and the industry relevant to the EXA2PRO use cases. At least two of the EXA2PRO use cases will have significant impact to the CO2 capture and to the
SDK4ED:Software Development toolKit for Energy optimization and technical Debt elimination
Type of action: RIA
Call identifier: H2020-ICT-2016-2017
The Software Engineering Group, of the Software and Data Engineering Laboratory, of the Department of Applied Informatics at the University of Macedonia, proudly announces its participation in the SDK4ED consortium.
The vision of SDK4ED is to minimize cost, time and complexity of low-energy software development processes, by providing tools for automatic optimization of both software quality and non-functional requirements such as energy efficiency, dependability and performance, with the capacity to tackle the interplay between design quality and run-time constraints.
SDK4ED aims to realise its vision through the following objectives: – establish a set of methods and tools for monitoring processes for early identification of design flaws, energy consumption indicators, and security vulnerabilities, with respect to the targeted hardware platform and non-functional requirements – estimate the costs and limitations associated to technical debt (TD) liabilities in the entire software stack – provide toolboxes for assessing project management decisions with respect to the choices of repaying TD, under the constraints imposed on energy consumption and security – deploy the envisaged solutions in three industry-driven distinctive but complementary use cases in the domains of airborne systems, healthcare, and automotive industry – Illustrate the importance and benefits introduced by proper TD management into low-energy software application development – train and consult the embedded software systems industry.
Through its envisaged toolboxes, SDK4ED will comprise a set of software programming add-ons for preventing the degradation of run-time qualities and especially energy consumption, while allowing for efficient measuring of the accumulated TD during the development of new low-energy computing software applications, including embedded systems and IoT products. The major expected impact of the proposed platform will be measured by the achieved improvement in productivity, the extent to which the envisaged tools will be adopted by the reference market and the minimisation of effort for adopting digital technologies into low-energy products and services.
Our paper “Forecasting Java Software Evolution Trends employing Network Models” has been accepted for publication in the IEEE Transactions on Software Engineering