[ https://issues.apache.org/jira/browse/AIRFLOW-6649?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17034360#comment-17034360 ]
Kamil Bregula commented on AIRFLOW-6649: ---------------------------------------- I agree, but I know that the internal implementation of Snowflake makes it the most efficient when it reads data from GCS/S3. Writing data directly is simply much slower. Snowflake can read data using multiple nodes when doing COPY INTO However, I support your idea, but I'm afraid that this might be problematic. I wonder about the operator dedicated to copying data using COPY INTO. It could start the copying process and then monitor the process and share logs directly in Airflow. We have a similar solution in the KubernetesPodOperator. > Google storage to Snowflake > --------------------------- > > Key: AIRFLOW-6649 > URL: https://issues.apache.org/jira/browse/AIRFLOW-6649 > Project: Apache Airflow > Issue Type: New Feature > Components: gcp, operators > Affects Versions: 1.10.6 > Reporter: nexoriv > Priority: Major > Labels: snowflake > > can someone share google storage to snowflake operator? -- This message was sent by Atlassian Jira (v8.3.4#803005)