From d9ce8ff757d9100ef0fdf83de1e4201be33cfbb7 Mon Sep 17 00:00:00 2001 From: Hongchao Deng Date: Mon, 18 May 2020 09:40:35 -0700 Subject: [PATCH] Update design/20200516-dependency.md Co-authored-by: Ryan Zhang <59209500+ryanzhang-oss@users.noreply.github.com> --- design/20200516-dependency.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/design/20200516-dependency.md b/design/20200516-dependency.md index 1f18341..dd51dfd 100644 --- a/design/20200516-dependency.md +++ b/design/20200516-dependency.md @@ -64,7 +64,7 @@ They have strict ordering: preprocess -> training -> serving ``` -Without the previous stage finished, the next couldn't start due to lack of data. +Each component has to wait for data to be finished processing by the previous stage. Today, ML frameworks including Tensorflow do not handle such dependency issue. They rely on external tools (e.g. Argo Workflow) to handle it. If we want to model them as Components in the same ApplicationConfiguration, the external tools could not help in this case.