Utilize este identificador para referenciar este registo: http://hdl.handle.net/10400.21/7181
Título: Optimizing memory transactions for large-scale programs
Autor: Carvalho, Fernando Miguel
Cachopo, João
Palavras-chave: Software Transactional Memory
Runtime optimizations
Concurrent programming
Data: 2016
Editora: Elsevier
Citação: CARVALHO, Fernando Miguel; CACHOPO, João - Optimizing memory transactions for large-scale programs. Journal of Parallel and Distributed Computing. ISSN 0743-7315. Vol. 89, (2016), pp. 13-24
Resumo: Even though Software Transactional Memory (STM) is one of the most promising approaches to simplify concurrent programming, current STM implementations incur significant overheads that render them impractical for many real-sized programs. The key insight of this work is that we do not need to use the same costly barriers for all the memory managed by a real-sized application, if only a small fraction of the memory is under contention—lightweight barriers may be used in this case. In this work, we propose a new solution based on an approach of adaptive object metadata (AOM) to promote the use of a fast path to access objects that are not under contention. We show that this approach is able to make the performance of an STM competitive with the best fine-grained lock-based approaches in some of the more challenging benchmarks.
Peer review: yes
URI: http://hdl.handle.net/10400.21/7181
DOI: 10.1016/j.jpdc.2015.12.001
ISSN: 0743-7315
Versão do Editor: http://ac.els-cdn.com/S0743731515002099/1-s2.0-S0743731515002099-main.pdf?_tid=03f19fe6-5276-11e7-9dad-00000aacb362&acdnat=1497605428_eb2a72741ee5568164dee5cd982bdf4b
Aparece nas colecções:ISEL - Eng. Elect. Tel. Comp. - Artigos

Ficheiros deste registo:
Ficheiro Descrição TamanhoFormato 
Optimizing_FMCarvalho_ADEETC.pdf1,77 MBAdobe PDFVer/Abrir    Acesso Restrito. Solicitar cópia ao autor!

FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Degois 

Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.