Big Data Meets Big Science.

The article focuses on the limitations of massively parallel computing in the analysis of data from next-generation scientific instruments. It states that powerful large-scale scientific instruments, such as the Large Synoptic Survey Telescope which is scheduled to go live in 2020, produce more data...

Descripción completa

Detalles Bibliográficos
Publicado en:Communications of the ACM Vol. 57; no. 7; pp. 13 - 16
Autor principal: Wright, Alex
Formato: Artículo
Publicado: Association for Computing Machinery Jul2014
Materias:
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:The article focuses on the limitations of massively parallel computing in the analysis of data from next-generation scientific instruments. It states that powerful large-scale scientific instruments, such as the Large Synoptic Survey Telescope which is scheduled to go live in 2020, produce more data than the most powerful massively parallel supercomputers can handle. It mentions that scientists are examining new methods of reducing datasets to a manageable size, including cloud-based computing and emerging frameworks like quantum computing, and talks about algorithmic and economic constraints facing large scale data analysis. Massachusetts Institute of Technology professor Scott Aaronson suggests that Moore's Law has effectively broken down.