Dataset of "Network Traffic Characteristics of Machine Learning Frameworks Under the Microscope" ...

Network traffic collection (PCAP) of three widely-used state-of-the-art Distributed Machine Learning (DML) frameworks (Tensorflow, Horovod, KungFu). The collection contains distributed training runs of four models (MobileNetV2, ResNet50, Resnet101, DenseNet201) with varying configurations of the fra...

Full description

Bibliographic Details
Main Authors: Zerwas, Johannes, Aykurt, Kaan, Schmid, Stefan, Blenk, Andreas
Format: Dataset
Language:English
Published: Technical University of Munich 2021
Subjects:
Distributed Machine Learning; Network Traffic Measurement
DML
Online Access:https://dx.doi.org/10.14459/2021mp1632489
https://mediatum.ub.tum.de/1632489
Description
Summary:Network traffic collection (PCAP) of three widely-used state-of-the-art Distributed Machine Learning (DML) frameworks (Tensorflow, Horovod, KungFu). The collection contains distributed training runs of four models (MobileNetV2, ResNet50, Resnet101, DenseNet201) with varying configurations of the frameworks. Varied parameters are the communication topology and backend, the distributed optimizer, the batch size and the packet loss in the network. ...

Similar Items