Does anybody have an idea how S3 Tables can achieve 3x performance compared to regular Iceberg tables on S3?
Does it work similar to the object-store file layout?
https://iceberg.apache.org/docs/latest/aws/#object-store-file-layout
#databs
Does it work similar to the object-store file layout?
https://iceberg.apache.org/docs/latest/aws/#object-store-file-layout
#databs
Comments
The comparison is against an uncompacted table in standard bucket. Presumably there's no speedup relative to a manually compacted table in a standard bucket.
Thanks for the explanation.