DNN learning jobs are common in today's clusters due to the advances in AI driven services such as machine translation and image recognition. The most critical phase of these jobs for model performance and learning cost is the tuning of hyperparameters. Existing approaches make use of techniques such as early stopping criteria to reduce the tuning impact on learning cost. However, these strategies do not consider the impact that certain hyperparameters and systems parameters have on training time. This paper presents PIPETUNE, a framework for DNN learning jobs that addresses the trade-offs between these two types of parameters. PipeTune takes advantage of the high parallelism and recurring characteristics of such jobs to minimize the learning cost via a pipelined simultaneous tuning of both hyper and system parameters. Our experimental evaluation using three different types of workloads indicates that PIPETUNE achieves up to 22.6% reduction and 1.7x speed up on tuning and training time, respectively. PIPETUNE not only improves performance but also lowers energy consumption up to 29%.
PipeTune: Pipeline Parallelism of Hyper and System Parameters Tuning for Deep Learning Clusters
Robert Birke;
2020-01-01
Abstract
DNN learning jobs are common in today's clusters due to the advances in AI driven services such as machine translation and image recognition. The most critical phase of these jobs for model performance and learning cost is the tuning of hyperparameters. Existing approaches make use of techniques such as early stopping criteria to reduce the tuning impact on learning cost. However, these strategies do not consider the impact that certain hyperparameters and systems parameters have on training time. This paper presents PIPETUNE, a framework for DNN learning jobs that addresses the trade-offs between these two types of parameters. PipeTune takes advantage of the high parallelism and recurring characteristics of such jobs to minimize the learning cost via a pipelined simultaneous tuning of both hyper and system parameters. Our experimental evaluation using three different types of workloads indicates that PIPETUNE achieves up to 22.6% reduction and 1.7x speed up on tuning and training time, respectively. PIPETUNE not only improves performance but also lowers energy consumption up to 29%.File | Dimensione | Formato | |
---|---|---|---|
2020 Middleware PipeTune- Pipeline Parallelism of Hyper and System Parameters Tuning for Deep Learning Clusters.pdf
Accesso riservato
Dimensione
2.77 MB
Formato
Adobe PDF
|
2.77 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.