Skip to main content

A Model-driven development framework for highly Parallel and EneRgy-Efficient computation supporting multi-criteria optimisation

Deliverables

Independent run-time energy support, and predictability, segregation and resilience mechanisms

Independent run-time energy support, and predictability, segregation and resilience mechanisms. The independent set of methods containing: (1) the dynamic energy management policies, (2) scheduling algorithms for managing computation and communication in a predictable way, (3) segregation mechanisms for safety/security support and (4) mechanisms for resilience support; and a report describing it.

First release of the meta parallel programming abstraction and the single-criterion performance-aware

First release of the meta parallel programming abstraction and the single-criterion performance- aware component. First released of the meta parallel programming abstraction applying single-criterion optimization model transformation; and the performance-aware single-criterion optimization component (not integrated with other optimization components) and a report describing it.

Single-criterion AMPERE ecosystem

Single-criterion AMPERE ecosystem. Synthesis tools, the software stack integrating run-time libraries, OS and hypervisor, addressing single-criterion optimisations; and a report describing it.

Single-criterion operating systems and hypervisor software

Single-criterion operating systems and hypervisor software. Preliminary version of: (1) the GPOS and RTOS supporting single-criterion run-time frameworks and (2) the PikeOS supporting singlecriterion run-time frameworks and OSs; and a user manual for installing and configuring them.

First release of the meta model-driven abstraction release

First release of the meta model-driven abstraction release. First release of the meta model-driven abstraction, upon which single-criterion optimization model transformations have been applied.

Progress Report for Technical Review

Progress report describing the project activities performed during the first 9 months of the project.

Communication and Dissemination Plan

Communication and Dissemination Plan. Description of the communication and dissemination strategy and the activities to be undertaken to achieve this.

Run-time Architecture

Run-time Architecture. Specification of the run-time mechanisms to be supported by AMPERE, describing the features and interfaces.

Model transformation requirements

Model transformation requirements. It contains: (1) Identification of WP2 requirements to capture functional and non-functional constraints, (2) analysis of state-of-the-art synthesis tools.

Multi-criteria optimization requirements

Multi-criteria optimization requirements. Identification of WP requirements

AMPERE ecosystem requirements and integration plan

AMPERE ecosystem requirements and integration plan. Identification of requirements,software components and tools common interface and detailed integration plan.

Project management plan and quality guidelines

Project management plan and quality guidelines. Description of internal management procedures, quality assurance process, and risk evaluation and internal communication tools and mechanisms.

Single-criterion energy optimisation framework, predictable execution models and software resilient techniques

Single-criterion energy optimisation framework, predictable execution models and software resilient techniques. Report describing: (1) energy-efficiency approaches, (2) the predictable execution models and (3) software resilient solutions for parallel execution, without considering the other constraints.

Analysis of functional safety aspects on single-criterion opti- mization and first release of the test bench suite

System models requirement and use case selection. Identification of: (1) use case constraints and requirements impose on other WPs, (2) functional safety and security aspects to be addressed.

Initial exploitation report

Initial exploitation report. Initial analysis of the exploitation context, business opportunities and exploitable results identifications, based on the requirements gathered at Phase 1.

Reference parallel heterogeneous hardware selection

Reference parallel heterogeneous hardware selection. Survey of the characteristics of analysed parallel heterogeneous platforms, illustrating the selected ones.

Initial communication and dissemination report

Initial communication and dissemination report. Report of the communication, dissemination and knowledge transfer activities until month 12.

Refined AMPERE ecosystem interfaces and integration plan

Refined AMPERE ecosystem interfaces and integration plan. Refined AMPERE common interface and integration plan.

Data Management Plan (DMP)

Data Management Plan. Analysis of the main elements of the data management policy with regard to all the datasets generated by the project.

Project management and collaboration tools

Project management and collaboration tools. Setup of the project management and collaboration tools, together with a small description report.

Publications

The OpenMP API for High Integrity Systems

Author(s): Michael Klemm, Eduardo Quiñones, Tucker Taft, Dirk Ziegenbein, Sara Royuela
Published in: ACM SIGAda Ada Letters, Issue 40/2, 2021, Page(s) 48-50, ISSN 1094-3641
DOI: 10.1145/3463478.3463480

EDF scheduling of real-time tasks on multiple cores

Author(s): Luca Abeni, Tommaso Cucinotta
Published in: ACM SIGAPP Applied Computing Review, Issue 20/2, 2020, Page(s) 5-18, ISSN 1559-6915
DOI: 10.1145/3412816.3412817

Dynamic partitioned scheduling of real-time tasks on ARM big.LITTLE architectures

Author(s): Agostino Mascitti, Tommaso Cucinotta, Mauro Marinoni, Luca Abeni
Published in: Journal of Systems and Software, Issue 173, 2021, Page(s) 110886, ISSN 0164-1212
DOI: 10.1016/j.jss.2020.110886

The AMPERE Project: : A Model-driven development framework for highly Parallel and EneRgy-Efficient computation supporting multi-criteria optimization

Author(s): Eduardo Quinones, Sara Royuela, Claudio Scordino, Paolo Gai, Luis Miguel Pinho, Luis Nogueira, Jan Rollo, Tommaso Cucinotta, Alessandro Biondi, Arne Hamann, Dirk Ziegenbein, Hadi Saoud, Romain Soulat, Bjorn Forsberg, Luca Benini, Gianluca Mando, Luigi Rucher
Published in: 2020 IEEE 23rd International Symposium on Real-Time Distributed Computing (ISORC), 2020, Page(s) 201-206
DOI: 10.1109/ISORC49007.2020.00042

OpenMP to CUDA graphs - a compiler-based transformation to enhance the programmability of NVIDIA devices

Author(s): Chenle Yu, Sara Royuela, Eduardo Quiñones
Published in: Proceedings of the 23th International Workshop on Software and Compilers for Embedded Systems, 2020, Page(s) 42-47
DOI: 10.1145/3378678.3391881

XPySom: High-Performance Self-Organizing Maps

Author(s): Riccardo Mancini, Antonio Ritacco, Giacomo Lanciano, Tommaso Cucinotta
Published in: 2020 IEEE 32nd International Symposium on Computer Architecture and High Performance Computing (SBAC-PAD), 2020, Page(s) 209-216
DOI: 10.1109/sbac-pad49847.2020.00037

A Synergistic Approach to Predictable Compilation and Scheduling on Commodity Multi-Cores

Author(s): Björn Forsberg, Maxim Mattheeuws, Andreas Kurth, Andrea Marongiu, Luca Benini
Published in: The 21st ACM SIGPLAN/SIGBED Conference on Languages, Compilers, and Tools for Embedded Systems, 2020, Page(s) 108-118
DOI: 10.1145/3372799.3394369

Automating the design flow under dynamic partial reconfiguration for hardware-software co-design in FPGA SoC

Author(s): Biruk Seyoum, Marco Pagani, Alessandro Biondi, Giorgio Buttazzo
Published in: Proceedings of the 36th Annual ACM Symposium on Applied Computing, 2021, Page(s) 481-490
DOI: 10.1145/3412841.3441928

A Toolchain to Verify the Parallelization of OmpSs-2 Applications

Author(s): Simone Economo, Sara Royuela, Eduard Ayguadé, Vicenç Beltran
Published in: Euro-Par 2020: Parallel Processing - 26th International Conference on Parallel and Distributed Computing, Warsaw, Poland, August 24–28, 2020, Proceedings, Issue 12247, 2020, Page(s) 18-33
DOI: 10.1007/978-3-030-57675-2_2

Static Analysis to Enhance Programmability and Performance in OmpSs-2

Author(s): Adrian Munera, Sara Royuela, Roger Ferrer, Raul Peñacoba, Eduardo Quiñones
Published in: High Performance Computing - ISC High Performance 2020 International Workshops, Frankfurt, Germany, June 21–25, 2020, Revised Selected Papers, Issue 12321, 2020, Page(s) 19-33
DOI: 10.1007/978-3-030-59851-8_2