SAVE
Self-Adaptive Virtualisation-Aware High-Performance/Low-Energy Heterogeneous System Architectures
| Coordinatore | POLITECNICO DI MILANO
Organization address
address: PIAZZA LEONARDO DA VINCI 32 contact info |
| Nazionalità Coordinatore | Italy [IT] |
| Totale costo | 4˙240˙704 € |
| EC contributo | 2˙930˙000 € |
| Programma | FP7-ICT
Specific Programme "Cooperation": Information and communication technologies |
| Code Call | FP7-ICT-2013-10 |
| Funding Scheme | CP |
| Anno di inizio | 2013 |
| Periodo (anno-mese-giorno) | 2013-09-01 - 2016-08-31 |
Partecipanti
| # | ||||
|---|---|---|---|---|
| 1 |
POLITECNICO DI MILANO
Organization address
address: PIAZZA LEONARDO DA VINCI 32 contact info |
IT (MILANO) | coordinator | 0.00 |
| 2 |
ARM LIMITED
Organization address
address: 110 FULBOURN ROAD contact info |
UK (CAMBRIDGE) | participant | 0.00 |
| 3 |
MAXELER TECHNOLOGIES LIMITED
Organization address
address: Down Place contact info |
UK (London) | participant | 0.00 |
| 4 |
STMICROELECTRONICS GRENOBLE 2 SAS
Organization address
address: RUE JULES HOROWITZ contact info |
FR (GRENOBLE) | participant | 0.00 |
| 5 |
TECHNOLOGICAL EDUCATIONAL INSTITUTE OF CRETE
Organization address
address: Ionni Athitaki & Gianni Kornarou contact info |
EL (HERAKLION) | participant | 0.00 |
| 6 |
UNIVERSITAET PADERBORN
Organization address
address: WARBURGER STRASSE contact info |
DE (PADERBORN) | participant | 0.00 |
| 7 |
VIRTUAL OPEN SYSTEMS SARL
Organization address
address: RUE LAKANAL contact info |
FR (GRENOBLE) | participant | 0.00 |
Mappa
Word cloud
Esplora la "nuvola delle parole (Word Cloud) per avere un'idea di massima del progetto.
Obiettivo del progetto (Objective)
The increasing availability of different kinds of processing resources in Heterogeneous System Architectures (HSA) associated with today's fast-changing, unpredictable workloads (e. g. of mobile or cloud-computing contexts), has propelled an interest towards self-adaptive systems that dynamically reorganise system resources to optimise for a given goal (e.g., performance, energy, reliability, resource utilisation). Hardware-assisted virtualisation is a key enabling technology for such HSAs; however it is available only for general-purpose CPUs, while heterogeneous resources, e.g. GPUs or FPGA-based dataflow engines (DFEs), currently do not support it. Thus, the performance/energy- efficiency benefits of these accelerators cannot be exploited in a self-adaptive HSAs that relies on virtualisation. The SAVE (Self-Adaptive Virtualisation-Aware High-Performance/Low-Energy Heterogeneous System Architectures) project will develop a stack of hardware, software and OS components that allow for deciding at run-time to execute tasks on the appropriate type of resource, based on the current system status/environment/application requirements. This objective is supported by two main innovations:n1. novel runtime OS components to manage the HSA by migrating task (or virtual machine) execution among CPUs, GPUs and DFEs, andn2. hardware-assisted virtualisation support for GPUs and DFEs.nThe project outcome will be an improved HSA with self-adaptation providing not only runtime reaction to changes, but also the means to dynamically achieve optimisation goals based on the current context. To this end, SAVE brings together experts from HPC and ES to benefit from high performance, power-efficient solutions, as well as self- adaptive fine-tuning of heterogeneous resources. SAVE technology will be demonstrated on development platforms that consist of state-of-the-art processor, network-on-chip, GPU, and DFE components using applications across both domains.