Scripting Multiple CPUs with Safe Data Sharing

Autor:	Roberto Ierusalimschy, Noemi Rodriguez, Alexandre Skyrme
Rok vydání:	2014
Předmět:	business.industry Programming language Non-lock concurrency control Computer science Concurrency Information sharing Software development computer.software_genre Data sharing Software Scripting language Multithreading Concurrent computing business Compiled language computer
Zdroj:	IEEE Software. 31:44-51
ISSN:	1937-4194 0740-7459
DOI:	10.1109/ms.2014.102
Popis:	Scripting languages are very popular and are being used to implement a wide range of applications. Meanwhile, multi-core processors are everywhere, from dektop computers to mobile devices, and concurrency has become the only means to improve performance. However, concurrent programming remains difficult and despite some interest in researching new concurrency models for compiled languages, the conventional concurrency support in scripting languages is still lacking. In this paper we are interested in \emph{safe data sharing for scripting languages}. We explore the landscape of conventional concurrency support in popular scripting languages and discuss some of its limitations. Then, we identify some emerging concurrency patterns in scripting languages, namely: no-default sharing, data ownership, futures and data immutability. Finally, we analyze some limitations in existing mechanisms and discuss how the patterns we identified can be used to provide safe data sharing in scripting languages.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::5e9e88327397535313382a100c7b9985 https://doi.org/10.1109/ms.2014.102 Zobrazit plný text záznamu