Michelle Blanc, M.Sc. commerce électronique. Marketing Internet, consultante, conférencière, auteure. 18 ans d'expérience

Le Parallelisme, la prochaine nouvelle affaire?

C’est via un twitt de Houssein (et aussi sur son blogue) que j’aie pris connaissance de la prochaine innovation possible de l’évolution du Web. On appelle ça le parallelisme et il semble que déjà, on prévoit une pénurie d’expertise dans ce domaine. Avis aux étudiants en informatique et aux chercheurs, de s’intéresser à ce domaine pourrait faire de vous des gens en demande.

Qu’est ce que le parallelisme? (fr.wikipedia.org)

La puissance de calcul de l'informatique, son potentiel en matière de modélisation, permettent d'explorer les domaines où dominent un grand nombre de phénomènes simultanés. L'informatique s'est d'ailleurs appliqué à elle-même ses capacités, les processus répétitifs pouvant partiellement se concevoir simultanément s'ils sont suffisamment indépendants. Le parallélisme désigne l'alternative aux processus séquentiels des algorithmes primitifs. Il repose sur des innovations d'ordre technique et matériel (architecture parallèle) autant que sur de nouveaux modes de programmation (traitements parallèles), voir réseau de neurones. D'un point de vue théorique, les automates cellulaires peuvent être considérés comme un modèle des comportements parallèles.

(en.wikipedia.org)

Parallel computing exists in several different forms: bit-level parallelism, instruction level parallelism, data parallelism, and task parallelism. It has been used for many years, mainly in high performance computing, but interest in it has become greater in recent years due to physical constraints preventing frequency scaling. Parallel computing has recently become the dominant paradigm in computer architecture, mainly in the form of multicore processors.[2]
Parallel computer programs are harder to write than sequential ones,[3] because concurrency introduces several new classes of potential software bugs, of which race conditions are the most common. Communication and synchronization between the different subtasks is typically one of the greatest barriers to getting good parallel program performance. In recent years, power consumption in parallel computers has also become a great concern.[4] The speed up of a program as a result of parallelization is given by Amdahl's law.

Et computingatscale qui y va de prédictions

• The move from sequential to parallel computing that is now underway will be as profound a change for the IT industry as the move to the web in the 1990s or the move to personal computers and workstations in the 1980s. If any one area is the “new new thing” for entrepreneurs and investors it’s parallelism. The drivers behind incredible opportunities in this area are simple, clear and unstoppable:
o All new computers are now parallel, and the scale of parallelism is growing exponentially.
o Only around 1% of the world’s software developers have any experience of parallel programming. The percentage that has significant experience in the design, analysis and implementation of highly parallel systems is perhaps as low as 0.01% (1 in 10,000).
o Every major hardware and software company in the business of delivering high performance solutions now has the difficult transition to parallelism as a major technical challenge.
This once-in-a-lifetime transition from sequential to parallel software promises a ten year bonanza for entrepreneurs and investors as new companies emerge and shake up the IT industry offering the tools, systems and applications for massively parallel computing.

 Imprimez ce billet

Article publié le lundi, 31 mars 2008 sous les rubriques Innovation, Technologies Internet et Web 2.0.

Vous pouvez suivre les commentaires sur cet article via ce fil RSS. Les commentaires sont présentement fermés.