Name: Parallel Computing Cluster for Solving Computational Problems in Data (en Inglés)
SKU: 66365111
Price: 65.75 USD
Availability: InStock
Author: Mutyala Rao, Chavala
ISBN: 9786209340734

portada Parallel Computing Cluster for Solving Computational Problems in Data (en Inglés)

Parallel Computing Cluster for Solving Computational Problems in Data (en Inglés)

Formato

Libro Físico

Editorial

LAP Lambert Academic Publishing

Autor

Mutyala Rao, Chavala

Idioma

Inglés

N° páginas

Encuadernación

Tapa Blanda

Dimensiones

22.9x15.2x0.3 cm

ISBN13

9786209340734

Parallel Computing Cluster for Solving Computational Problems in Data (en Inglés)

Mutyala Rao, Chavala (Autor) · LAP Lambert Academic Publishing · Tapa Blanda

Libro Nuevo Importado

Envío: 26 a 33 días háb.

$ 109.59$ 65.75

-40%

Costos de importación incluídos en el precio ✅

Libro Nuevo

Quedan 10 unidades

$ 65.75

Llega entre el 18 Nov y el 01 Dic a Quito, Pichincha. Seleccionar ubicación

Agregar a lista de deseos

Reseña del libro "Parallel Computing Cluster for Solving Computational Problems in Data (en Inglés)"

Thisbook tackles a critical bottleneck in large-scale AI: the slow and communication-heavy training of massive Deep Neural Networks (DNNs) on multi-GPU systems. It addresses the trade-off between two main parallelization methods. Data parallelism suffers from severe communication overhead for large models, while pipelined model parallelism (like PipeDream) offers up to 8.91x speedup for large Fully Connected/Recurrent Neural Networks but causes "weight staleness," degrading model accuracy. To resolve this, the paper introduces SpecTrain, a novel technique. SpecTrain uses the momentum from optimizers to predict future weight updates, allowing pipelined computation with accurate, non-stale weights. This enables the high GPU utilization and speed of pipelining while maintaining the training robustness and final accuracy of synchronous methods.