Loading…
Back To Schedule
Wednesday, May 22 • 15:55 - 16:30
Large Scale Distributed Deep Learning with Kubernetes Operators - Yuan Tang, Ant Financial & Yong Tang, MobileIron

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

Tweet Share
Feedback form is now closed.
The focus of this talk is the usage of Kubernetes operators to manage and automate training process for machine learning tasks. Two open source Kubernetes operators, tf-operator and mpi-operator, will be discussed. Both operators manage training jobs for TensorFlow but they have different distribution strategies. The tf-operator fits the parameter server distribution strategy which has a centralized parameter server for coordination. The mpi-operator, on the other hand, utilize MPI allreduce primitive implementation. While the parameter server strategy requires a right ratio of CPU (for parameter servers) and GPU (for workers) to reach network-optimal, the all reduce distribution might be easier to optimize network cost. We will share our performance numbers in out talk for comparison of those two operators.
Speakers
avatar for Yong Tang

Yong Tang

Director of Engineering, Ivanti, Inc
Yong Tang is director of engineering at Ivanti. He is a core maintainer of CoreDNS and contributes to many container, cloud-native, and machine learning projects for the open source community. In addition to CoreDNS, he is a maintainer of Docker/Moby. He is also a maintainer and SIG... Read More →
avatar for Yuan Tang

Yuan Tang

Founding Engineer, Akuity
Yuan is currently a founding engineer at Akuity. Previously he was a senior software engineer at Alibaba Group, building AI infrastructure and AutoML platform. He's a PMC member of XGBoost and Apache MXNet, co-chair of Kubeflow, committer of TensorFlow, Argo, Couler, and ElasticDL... Read More →

KubeCon Barcelona Deep Learning pdf
Wednesday May 22, 2019 15:55 - 16:30 CEST
Hall 8.0 D2
  Machine Learning + Data