Skip to content

feat: enable native Iceberg reader by default#3819

Draft
andygrove wants to merge 2 commits intoapache:mainfrom
andygrove:enable-iceberg-native-default
Draft

feat: enable native Iceberg reader by default#3819
andygrove wants to merge 2 commits intoapache:mainfrom
andygrove:enable-iceberg-native-default

Conversation

@andygrove
Copy link
Copy Markdown
Member

Which issue does this PR close?

Closes #.

Rationale for this change

The native Iceberg reader has matured enough to be enabled by default, removing the need for users to explicitly opt in via configuration.

What changes are included in this PR?

  • Change default value of spark.comet.scan.icebergNative.enabled from false to true
  • Update user-guide Iceberg documentation to reflect the new default and remove the config from shell examples
  • Update benchmark README to note the feature is enabled by default

How are these changes tested?

Existing Iceberg test suites cover the native reader. This change only flips the default; no new behavior is introduced.

Change the default value of spark.comet.scan.icebergNative.enabled
from false to true, and update documentation to reflect this change.
@andygrove andygrove marked this pull request as ready for review March 27, 2026 21:37
@andygrove andygrove requested a review from mbutrovich March 27, 2026 21:37
@mbutrovich
Copy link
Copy Markdown
Contributor

I would say this issue is blocking for this PR: #3856.

@andygrove andygrove marked this pull request as draft March 31, 2026 17:24
@@ -140,7 +140,7 @@ object CometConf extends ShimCometConf {
"Iceberg tables are read directly through native execution, bypassing Spark's " +
"DataSource V2 API for better performance.")
Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

https://datafusion.apache.org/comet/user-guide/latest/iceberg.html says that Iceberg support is experimental. Some other config entries mention that they are experimental.

Suggested change
"DataSource V2 API for better performance.")
"DataSource V2 API for better performance. This feature is experimental.")

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants