H2020Индивидуална стипендия2016–2018

DURO · Deep-memory Ubiquity, Reliability and Optimization

„Хоризонт 2020“ — Действия „Мария Склодовска-Кюри“

Период: 2016-10-01 → 2018-09-30
Финансиране от ЕС: 170 122 €
Участници: 1
Схема: MSCA-IF-EF-ST

Линиите свързват координатора с партньорите.

Накратко на български

Стратегиите за управление на паметта в суперкомпютрите се изследват, за да се оптимизира движението на данни при симулации на климата или галактики. Това помага за повишаване на производителността и надеждността на машините, които обработват огромни масиви от информация.

Този кратък обзор е генериран от изкуствен интелект

Кратко обяснение, генерирано от езиков модел по текста на CORDIS. Оригиналът е по-долу.

Резултати накратко

DURO: Deep-memory Ubiquity, Reliability and Optimization

Supercomputers are extremely important for scientific progress. Scientists use large supercomputers to model a large variety of phenomena, from sub-atomic particle behavior to the formation of galaxies, passing by a large number of industrial applications. Understanding climate change, mental disorders and new materials are just some of the many applications that require extremely large and long simulation in supercomputers. High-performance computing has been growing exponentially for the last decades, and they still need to increase their computational power by several orders of magnitude in order to run the most ambitious scientific simulations on them. However, increasing computational power must come together with an increase in data availability, that is to say fast data movement between processors and storage. To this end, deep memory hierarchies have come to exist, with several storage devices offering multiple trade-offs between capacity, latency, bandwidth, resilience, among others. Using those deep memory hierarchies efficiently is mandatory to extract all the computational power from future extreme scale machines. The objectives of this project is to design strategies to increase simultaneously the fault tolerance and the performance of scientific applications in deep memory hierarchies. For that it is important to analyze the applications usually running on large supercomputers and extract features that could map well with the memory limitations existing in the systems.

Текст от CORDIS, на английски · Данни: CORDIS, © Европейски съюз

Цел на проекта

High performance computing has changed the way scientists make discoveries and is driving industrial innovation. From the simulation of the origins of the universe, to the optimization of wind turbine placement; supercomputers are helping us to change the world and improve the conditions of future generations. The next generation of European extreme scale computers brings new opportunities but also imposes new challenge: they need to be an order of magnitude more energy efficient and they need to be reliable so that no data is lost in the presence of failures. Novel deep-memory hierarchies offer an alternative to achieve these resilience and efficiency goals. Unfortunately, it is unclear how the system should utilize these cutting-edge hardware devices. The objective of this project is to build an abstraction layer between the new hybrid memory hardware and the scientific application, providing an easy way to leverage the features of the hardware while maintaining high energy efficiency and strong reliability. Barcelona Supercomputing Centre is an ideal place to carry out this research because of its top-level researchers. In particular, the fellowship will be supervised by Dr. Osman Unsal, who has a long outstanding experience supervising European projects. The combination of such an important project with the high quality training of the host institution represents the best career opportunity for the candidate to expand and solidify his research experience.

Оригинален текст от CORDIS (на английски).

Участници

BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACION · BARCELONAКоординаторИспания