ApsaraDB for ClickHouse:Release details of ApsaraDB for ClickHouse Enterprise Edition V24.8

New features

• ASOF JOIN supports the full_sorting_join algorithm, improving the efficiency of sorting-based JOIN operations.

• The tuple function is extended to support queries that use named tuples.

• New functions include tupleNames, fuzzQuery, percent_rank, changeYear, changeMonth, changeDay, changeHour, changeMinute, changeSecond, and groupConcat.

• Directory operations are simplified. The file table function automatically appends the asterisk wildcard character (*) to directory paths.

• The readWKTLineString function reads MULTILINESTRING data in the WKT format.

• Parquet files support a new configuration to enable or disable page index writing.

• Hive-style partitioning supports File, URL, S3, AzureBlobStorage, and HDFS storage engines for efficient querying and management of large datasets.

• The printf function supports Spark-compatible formatted output.

• The S3 table engine adds a virtual column _etag, improving processing efficiency for S3 data.

• The query cache supports namespaces, allowing identical queries with different tags to be treated as separate cache entries.

• The system.error_log system table records historical error information from the system.errors table and periodically flushes data to disk.

• The --memory-usage client option monitors memory usage in non-interactive mode.

Performance optimization

• Binary serialization of the Variant data type is optimized by enabling compact mode, reducing redundant data storage.

• S3 table engine data access performance is improved for faster queries.

Breaking changes

• The CREATE MATERIALIZED VIEW ... ENGINE Replicated*MergeTree POPULATE AS SELECT syntax cannot be used in replicated databases.

• The alternative LowCardinality syntax using the WithDictionary suffix is deprecated. Update your table schemas to use the standard LowCardinality data type.

Fixed issues

• Fixed an issue where the Buffer storage engine used with destination distributed tables caused query failures, especially when the same table was referenced multiple times in a query.

• Fixed an issue where Gamma-based random distribution functions (Chi-Squared, Student, and Fisher) caused extended computation or infinite loops when parameters had near-zero negative values.

• Fixed an issue where the arrayWithConstant function had performance problems with large arrays. Array size is now limited to 1 GB.

• Fixed format issues in low-level memory management to enhance the stability of memory allocations.

• Fixed compatibility issues between the Buffer storage engine and destination distributed tables.

• Fixed optimization errors when the IS NULL condition was used in join operations.