docs

dbms/src/Core/Settings.h

@@ -108,9 +108,9 @@ struct Settings : public SettingsCollection<Settings>
     M(SettingBool, distributed_group_by_no_merge, false, "Do not merge aggregation states from different servers for distributed query processing - in case it is for certain that there are different keys on different shards.") \
     M(SettingBool, optimize_skip_unused_shards, false, "Assumes that data is distributed by sharding_key. Optimization to skip unused shards if SELECT query filters by sharding_key.") \
     \
-    M(SettingBool, input_format_parallel_parsing, true, "Enable parallel parsing for several data formats (JSON, TSV, TKSV, Values, CSV).") \
-    M(SettingUInt64, max_threads_for_parallel_reading, 10, "The maximum number of threads to parallel reading. By default, it is set to max_threads.") \
-    M(SettingUInt64, min_chunk_size_for_parallel_reading, (1024 * 1024), "The minimum chunk size in bytes, which each thread tries to parse under mutex in parallel reading.") \
+    M(SettingBool, input_format_parallel_parsing, true, "Enable parallel parsing for several data formats (JSONEachRow, TSV, TKSV, CSV).") \
+    M(SettingUInt64, max_threads_for_parallel_parsing, 10, "The maximum number of threads to parallel parsing.") \
+    M(SettingUInt64, min_chunk_size_for_parallel_parsing, (1024 * 1024), "The minimum chunk size in bytes, which each thread will parse in parallel.") \
     \
     M(SettingUInt64, merge_tree_min_rows_for_concurrent_read, (20 * 8192), "If at least as many lines are read from one file, the reading can be parallelized.") \
     M(SettingUInt64, merge_tree_min_bytes_for_concurrent_read, (24 * 10 * 1024 * 1024), "If at least as many bytes are read from one file, the reading can be parallelized.") \

dbms/src/DataStreams/ParallelParsingBlockInputStream.h

@@ -16,8 +16,8 @@ namespace DB
 /**
  * ORDER-PRESERVING parallel parsing of data formats.
  * It splits original data into chunks. Then each chunk is parsed by different thread.
- * The number of chunks equals to max_threads_for_parallel_reading setting.
- * The size of chunk is equal to min_chunk_size_for_parallel_reading setting.
+ * The number of chunks equals to max_threads_for_parallel_parsing setting.
+ * The size of chunk is equal to min_chunk_size_for_parallel_parsing setting.
  */
 class ParallelParsingBlockInputStream : public IBlockInputStream
 {

dbms/src/Formats/FormatFactory.cpp

@@ -124,12 +124,12 @@ BlockInputStreamPtr FormatFactory::getInput(
         row_input_format_params.max_execution_time = settings.max_execution_time;
         row_input_format_params.timeout_overflow_mode = settings.timeout_overflow_mode;
 
-        size_t max_threads_to_use = settings.max_threads_for_parallel_reading;
+        size_t max_threads_to_use = settings.max_threads_for_parallel_parsing;
         if (!max_threads_to_use)
             max_threads_to_use = settings.max_threads;
 
         auto params = ParallelParsingBlockInputStream::InputCreatorParams{sample, context, row_input_format_params, format_settings};
-        ParallelParsingBlockInputStream::Builder builder{buf, input_getter, params, file_segmentation_engine, max_threads_to_use, settings.min_chunk_size_for_parallel_reading};
+        ParallelParsingBlockInputStream::Builder builder{buf, input_getter, params, file_segmentation_engine, max_threads_to_use, settings.min_chunk_size_for_parallel_parsing};
         return std::make_shared<ParallelParsingBlockInputStream>(builder);
     }
 

docs/en/operations/settings/settings.md

@@ -980,4 +980,25 @@ Lower values mean higher priority. Threads with low `nice` priority values are e
 
 Default value: 0.
 
+## input_format_parallel_parsing
+
+- Type: bool
+- Default value: True
+
+Enable order-preserving parallel parsing of data formats such as JSONEachRow, TSV, TKSV and CSV. Reading will be single threaded and parsing will be multithreaded.
+
+## max_threads_for_parallel_parsing
+
+- Type: unsigned int
+- Default value: 10
+
+The maximum number of threads for order-preserving parallel parsing of data formats.
+
+## min_chunk_size_for_parallel_parsing
+
+- Type: unsigned int
+- Default value: 1024 * 1024
+
+The minimum chunk size in bytes, which each thread will parse in parallel. By default it equals to one megabyte.
+
 [Original article](https://clickhouse.yandex/docs/en/operations/settings/settings/) <!-- hide -->