Issues

• hyper: "request exceeding maximum allowed size for request" when reading a Parquet file with large row groups

  ClickHouse/ClickBench#3294 · Filter by author alexey-milovidov opened on Nov 13, 2026

• umbra: the server crashes while loading a billion-row Parquet file with a large footer

  ClickHouse/ClickBench#3292 · Filter by author alexey-milovidov opened on Nov 13, 2026

• hyper: a Parquet file with a footer over 16 MiB is refused

  ClickHouse/ClickBench#3291 · Filter by author alexey-milovidov opened on Nov 13, 2026
  1

• CedarDB can't work on large tables.

  ClickHouse/ClickBench#3288 · Filter by author alexey-milovidov opened on Nov 13, 2026
  1

• suggest add an auto tranlsate feature for playground

  ClickHouse/ClickBench#2650 · Filter by author l1t1 opened on Oct 17, 2026

• doris-parquet-single fails before measurement on c6a.xlarge: FE cannot commit its 8 GiB initial heap

  ClickHouse/ClickBench#2208 · Filter by author KazukiKandaKK opened on Sep 25, 2026
  2

• Consider removing partitioned variants of Parquet, Data Lake, Vortex

  ClickHouse/ClickBench#2146 · Filter by author alexey-milovidov opened on Sep 21, 2026
  1

• ClickBench+

  ClickHouse/ClickBench#1488 · Filter by author 4ertus2 opened on Aug 19, 2026
  1

• Refresh CHYT benchmark results

  close in a month if not activehelp wanted
  ClickHouse/ClickBench#1459 · Filter by author edlimonov opened on Aug 18, 2026

• Versions benchmark: Run TPC-H and TPC-DS queries in standard-compatible mode

  close in a month if not active
  ClickHouse/ClickBench#1322 · Filter by author rschu1ze opened on Aug 11, 2026

• Add Apache Kylin

  ClickHouse/ClickBench#1089 · Filter by author alexey-milovidov opened on Jul 25, 2026

• Add SynnoDB

  ClickHouse/ClickBench#960 · Filter by author michaelkoepf opened on Jun 30, 2026
  3

• concurrent_qps measures swap-thrashing, not throughput, for single-process engines (introduced in #860)

  ClickHouse/ClickBench#946 · Filter by author george-larionov opened on Jun 12, 2026
  6

  

• Some, but not all scripts report "load time" for stateless engines (time to download parquet and sync)

  ClickHouse/ClickBench#944 · Filter by author alamb opened on Jun 1, 2026

• ClickHouse: Cold runs preload data into memory before the timer

  ClickHouse/ClickBench#941 · Filter by author pfent opened on May 27, 2026
  1

• Methodology: 3 hot iterations is insufficient for JVM-based engines

  ClickHouse/ClickBench#934 · Filter by author puzpuzpuz opened on May 19, 2026
  3

• Add FrigateBird

  ClickHouse/ClickBench#809 · Filter by author alexey-milovidov opened on Mar 4, 2026

• "System & Machine" column should display sw version number(s)

  ClickHouse/ClickBench#639 · Filter by author DBguru opened on Oct 9, 2025
  5

• Add Yellowbrick to ClickBench (someone promised to do it, but didn't)

  ClickHouse/ClickBench#628 · Filter by author alexey-milovidov opened on Sep 17, 2025
  4

• MariaDB ColumnStore eats all memory and makes the machine unresponsive

  ClickHouse/ClickBench#545 · Filter by author alexey-milovidov opened on Jul 12, 2025
  8

• DuckDB: failed with OOM killer

  ClickHouse/ClickBench#542 · Filter by author alexey-milovidov opened on Jul 12, 2025
  1

• Doris、Starrocks、SelectDB can't load data

  ClickHouse/ClickBench#541 · Filter by author alexey-milovidov opened on Jul 12, 2025
  3

• Snowflake CSV - use parallel scan settings for uncompressed CSV

  ClickHouse/ClickBench#408 · Filter by author sfc-gh-xhuang opened on Jun 17, 2025
  1

• Add Turso (it is unbelievably slow) (it also does not work)

  ClickHouse/ClickBench#336 · Filter by author alexey-milovidov opened on Mar 23, 2025
  11

• InfluxDB benchmarks

  ClickHouse/ClickBench#186 · Filter by author paxw-panevo opened on May 2, 2024
  2