You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@dolphinscheduler.apache.org by jieguang_zhou <ji...@163.com> on 2022/03/14 06:22:49 UTC

回复:[Feature-Request] Introduce the task plugin of TonY intodolphinscheduler

Hi Junfan
That's a good suggestion. We're going to do Mlops-related scenes.


在 2022年3月14日 13:59,Junfan Zhang<ju...@outlook.com> 写道:


Hi the community Do you mind introducing the task plugin of TonY, which is a framework to natively run deep learning jobs on Apache Hadoop. It currently supports TensorFlow, PyTorch, MXNet and Horovod. TonY enables running either single node or distributed training as a Hadoop application. This native connector, together with other TonY features, aims to run machine learning  jobs reliably and flexibly.I think if can be supported into dolphinscheduler, it will make people reuse the existing on-perm Hadoop cluster resources and dolphinscheduler will also be a new workflow engine that covers the machine learning lifecycle jobs. Feel free to discuss if having any ideas. Best Junfan Zhang

Re: 回复:[Feature-Request] Introduce the task plugin of TonY intodolphinscheduler

Posted by Junfan Zhang <ju...@outlook.com>.
Hi

Thanks for your quick reply. So glad to hear that.

Sorry. i forget to attach the TonY website: https://tony-project.ai/

On Mar 14, 2022, 2:23 PM +0800, jieguang_zhou <ji...@163.com>, wrote:
> Hi Junfan
> That's a good suggestion. We're going to do Mlops-related scenes.
>
>
> 在 2022年3月14日 13:59,Junfan Zhang<ju...@outlook.com> 写道:
>
>
> Hi the community Do you mind introducing the task plugin of TonY, which is a framework to natively run deep learning jobs on Apache Hadoop. It currently supports TensorFlow, PyTorch, MXNet and Horovod. TonY enables running either single node or distributed training as a Hadoop application. This native connector, together with other TonY features, aims to run machine learning jobs reliably and flexibly.I think if can be supported into dolphinscheduler, it will make people reuse the existing on-perm Hadoop cluster resources and dolphinscheduler will also be a new workflow engine that covers the machine learning lifecycle jobs. Feel free to discuss if having any ideas. Best Junfan Zhang