commits  

Messages by Thread

• • Re: [PR] perf(spark): set scan-constant parquet read keys once per scan [hudi] via GitHub
  • Re: [PR] perf(spark): set scan-constant parquet read keys once per scan [hudi] via GitHub
  • Re: [PR] perf(spark): set scan-constant parquet read keys once per scan [hudi] via GitHub
• [I] Spark parquet base file reads convert and compare the full footer schema for every file [hudi] via GitHub
• (hudi-rs) branch ci/minio-chainguard deleted (was 6caa4fd6) vhs
• Re: [PR] feat(index): add RaBitQ encoder and multibit-aware scorer contract [hudi] via GitHub
  • Re: [PR] feat(index): add RaBitQ encoder and multibit-aware scorer contract [hudi] via GitHub
  • Re: [PR] feat(index): add RaBitQ encoder and multibit-aware scorer contract [hudi] via GitHub
  • Re: [PR] feat(index): add RaBitQ encoder and multibit-aware scorer contract [hudi] via GitHub
  • Re: [PR] feat(index): add RaBitQ encoder and multibit-aware scorer contract [hudi] via GitHub
  • Re: [PR] feat(index): add RaBitQ encoder and multibit-aware scorer contract [hudi] via GitHub
• [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
  • Re: [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
  • Re: [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
  • [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
  • Re: [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
  • Re: [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
  • Re: [PR] ci: pull MinIO images from Chainguard [hudi-rs] via GitHub
• [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
  • Re: [PR] fix(tools): read archived clean plans and metadata [hudi] via GitHub
• [I] Index lookup tasks ship the table and reload the timeline on executors [hudi] via GitHub
• [I] Table service tasks ship the table, and bootstrap listing ignores the job's Hadoop configuration [hudi] via GitHub
• [PR] perf(core): keep the table out of table service tasks and fix the bootstrap listing configuration [hudi] via GitHub
  • Re: [PR] perf(core): keep the table out of table service tasks and fix the bootstrap listing configuration [hudi] via GitHub
  • Re: [PR] perf(core): keep the table out of table service tasks and fix the bootstrap listing configuration [hudi] via GitHub
  • Re: [PR] perf(core): keep the table out of table service tasks and fix the bootstrap listing configuration [hudi] via GitHub
  • Re: [PR] perf(core): keep the table out of table service tasks and fix the bootstrap listing configuration [hudi] via GitHub
• [I] Metadata table lookup tasks ship the whole table metadata and read the timeline [hudi] via GitHub
• [PR] perf(metadata): keep table state and per-task metadata I/O out of metadata index record generation [hudi] via GitHub
  • Re: [PR] perf(metadata): keep table state and per-task metadata I/O out of metadata index record generation [hudi] via GitHub
  • Re: [PR] perf(metadata): keep table state and per-task metadata I/O out of metadata index record generation [hudi] via GitHub
  • Re: [PR] perf(metadata): keep table state and per-task metadata I/O out of metadata index record generation [hudi] via GitHub
  • Re: [PR] perf(metadata): keep table state and per-task metadata I/O out of metadata index record generation [hudi] via GitHub
• [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
  • Re: [PR] perf(metadata): stop shipping the table metadata to metadata table lookup tasks [hudi] via GitHub
• [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
  • Re: [PR] perf(index): keep tables and timeline reloads out of index lookup tasks [hudi] via GitHub
• [I] Metadata index record generation tasks carry table state and repeat metadata table I/O [hudi] via GitHub
• (hudi-rs) branch ci/minio-chainguard created (now 6caa4fd6) vhs
• [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
  • Re: [PR] perf(spark): decide the vectorized parquet reader per scan without mutating the session conf [hudi] via GitHub
• [I] Spark reads flip the session-wide vectorized Parquet reader flag, slowing later scans [hudi] via GitHub
• (hudi) branch master updated (8d3b06570c58 -> ead194cce548) sivabalan
• (hudi) branch bot/trino-pin updated (49c216c83e5f -> 328266a6418a) github-bot
  • (hudi) 01/01: chore(trino): advance trino master pin to b90cad6ffc48 github-bot
• [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
  • Re: [PR] perf(flink,hive): read splits with the table state captured at planning [hudi] via GitHub
• [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
  • Re: [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
  • Re: [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
  • Re: [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
  • Re: [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
  • Re: [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
  • Re: [PR] perf(spark,flink): resolve schema-on-read file schemas from a history loaded once per query [hudi] via GitHub
• [PR] perf(spark): trim the upsert partitioner, write stage and record creation task payloads [hudi] via GitHub
  • Re: [PR] perf(spark): trim the upsert partitioner, write stage and record creation task payloads [hudi] via GitHub
  • Re: [PR] perf(spark): trim the upsert partitioner, write stage and record creation task payloads [hudi] via GitHub
  • Re: [PR] perf(spark): trim the upsert partitioner, write stage and record creation task payloads [hudi] via GitHub
  • Re: [PR] perf(spark): trim the upsert partitioner, write stage and record creation task payloads [hudi] via GitHub
• [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
  • Re: [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
  • Re: [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
  • Re: [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
  • Re: [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
  • Re: [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
  • Re: [PR] perf(trino): read file groups on workers from table state shipped on the handle [hudi] via GitHub
• [I] Spark upsert partitioner, write stage and record creation tasks ship state they never read [hudi] via GitHub
• [I] Schema-on-read resolves each base file's schema from .hoodie on executors [hudi] via GitHub
• [I] Trino workers build a meta client for every split with log files [hudi] via GitHub
• [I] Flink and Hive readers build a meta client for every split [hudi] via GitHub
• [I] Reduce per-task executor overhead in Hudi 1.x [hudi] via GitHub
  • Re: [I] Reduce per-task executor overhead in Hudi 1.x [hudi] via GitHub
• [I] Spark file group reader ships the file format and meta client to every task [hudi] via GitHub
  • Re: [I] Spark file group reader ships the file format and meta client to every task [hudi] via GitHub
• [I] File group reader loads the timeline on executors for tables before version 8 [hudi] via GitHub
  • Re: [I] File group reader loads the timeline on executors for tables before version 8 [hudi] via GitHub
• [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
  • Re: [PR] fix(metadata): skip the meta folder when listing partitions from the file system [hudi] via GitHub
• [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
  • Re: [PR] perf(spark): stop per-file Hadoop configuration parsing and copies on the parquet read path [hudi] via GitHub
• [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub
  • Re: [PR] refactor(core): read file groups from a table state instead of a meta client [hudi] via GitHub

• Earlier messages
• Later messages