Different configuration parameters take effect in the following three ways:
Only allowed to be modified in first start up: Can't be modified after first start, otherwise the ConfigNode/DataNode cannot start.
After restarting system: Can be modified after the ConfigNode/DataNode first start, but take effect after restart.
hot-load: Can be modified while the ConfigNode/DataNode is running, and trigger through sending the command(sql) load configuration to the IoTDB server by client or session.
The number of SchemaRegionGroups that each Database has when using the CUSTOM-SchemaRegionGroup extension policy. The least number of SchemaRegionGroups that each Database has when using the AUTO-SchemaRegionGroup extension policy.
Type
int
Default
1
Effective
After restarting system
schema_region_per_data_node
Name
schema_region_per_data_node
Description
The maximum number of SchemaRegion expected to be managed by each DataNode
Type
double
Default
1.0
Effective
After restarting system
data_region_group_extension_policy
Name
data_region_group_extension_policy
Description
The extension policy of DataRegionGroup
Type
string
Default
AUTO
Effective
After restarting system
default_data_region_group_num_per_database
Name
default_data_region_group_num_per_database
Description
The number of DataRegionGroups that each Database has when using the CUSTOM-DataRegionGroup extension policy. The least number of DataRegionGroups that each Database has when using the AUTO-DataRegionGroup extension policy.
Type
int
Default
1
Effective
After restarting system
data_region_per_processor
Name
data_region_per_processor
Description
The maximum number of DataRegion expected to be managed by each processor
Type
double
Default
1.0
Effective
After restarting system
enable_data_partition_inherit_policy
Name
enable_data_partition_inherit_policy
Description
Whether to enable the DataPartition inherit policy
Type
Boolean
Default
false
Effective
After restarting system
leader_distribution_policy
Name
leader_distribution_policy
Description
The policy of cluster RegionGroups' leader distribution
Type
String
Default
MIN_COST_FLOW
Effective
After restarting system
enable_auto_leader_balance_for_ratis
Name
enable_auto_leader_balance_for_ratis_consensus
Description
Whether to enable auto leader balance for Ratis consensus protocol
Type
Boolean
Default
false
Effective
After restarting system
enable_auto_leader_balance_for_iot_consensus
Name
enable_auto_leader_balance_for_iot_consensus
Description
Whether to enable auto leader balance for IoTConsensus protocol
Memory Allocation Ratio: StorageEngine, QueryEngine, SchemaEngine, Consensus and Free Memory
Type
Ratio
Default
3:3:1:1:2
Effective
After restarting system
schema_memory_allocate_proportion
Name
schema_memory_allocate_proportion
Description
Schema Memory Allocation Ratio: SchemaRegion, SchemaCache, PartitionCache and LastCache
Type
Ratio
Default
5:3:1:1
Effective
After restarting system
storage_engine_memory_proportion
Name
storage_engine_memory_proportion
Description
Memory allocation ratio in StorageEngine: Write, Compaction
Type
Ratio
Default
8:2
Effective
After restarting system
write_memory_proportion
Name
write_memory_proportion
Description
Memory allocation ratio in writing: Memtable, TimePartitionInfo
Type
Ratio
Default
19:1
Effective
After restarting system
concurrent_writing_time_partition
Name
concurrent_writing_time_partition
Description
This config decides how many time partitions in a database can be inserted concurrently For example, your partitionInterval is 86400 and you want to insert data in 5 different days,
Type
int32
Default
1
Effective
After restarting system
primitive_array_size
Name
primitive_array_size
Description
primitive array size (length of each array) in array pool
Type
Int32
Default
64
Effective
After restart system
chunk_metadata_size_proportion
Name
chunk_metadata_size_proportion
Description
size proportion for chunk metadata maintains in memory when writing tsfile
Type
Double
Default
0.1
Effective
After restart system
flush_proportion
Name
flush_proportion
Description
Ratio of write memory for invoking flush disk, 0.4 by default If you have extremely high write load (like batch=1000), it can be set lower than the default value like 0.2
Type
Double
Default
0.4
Effective
After restart system
buffered_arrays_memory_proportion
Name
buffered_arrays_memory_proportion
Description
Ratio of write memory allocated for buffered arrays
Type
Double
Default
0.6
Effective
After restart system
reject_proportion
Name
reject_proportion
Description
Ratio of write memory for rejecting insertion
Type
Double
Default
0.8
Effective
After restart system
write_memory_variation_report_proportion
Name
write_memory_variation_report_proportion
Description
if memory cost of data region increased more than proportion of allocated memory for write, report to system
Type
Double
Default
0.001
Effective
After restarting system
check_period_when_insert_blocked
Name
check_period_when_insert_blocked
Description
when an inserting is rejected, waiting period (in ms) to check system again
Type
Int32
Default
50
Effective
After restart system
io_task_queue_size_for_flushing
Name
io_task_queue_size_for_flushing
Description
size of ioTaskQueue. The default value is 10
Type
Int32
Default
10
Effective
After restart system
enable_query_memory_estimation
Name
enable_query_memory_estimation
Description
If true, we will estimate each query's possible memory footprint before executing it and deny it if its estimated memory exceeds current free memory
Type
bool
Default
true
Effective
hot-load
partition_cache_size
Name
partition_cache_size
Description
The max num of partition info record cached on DataNode.
Schema engine mode, supporting Memory and PBTree modes; PBTree mode support evict the timeseries schema temporarily not used in memory at runtime, and load it into memory from disk when needed. This parameter must be the same on all DataNodes in one cluster.
Type
string
Default
Memory
Effective
Only allowed to be modified in first start up
mlog_buffer_size
Name
mlog_buffer_size
Description
size of log buffer in each metadata operation plan(in byte)
Type
int32
Default
1048576
Effective
After restart system
sync_mlog_period_in_ms
Name
sync_mlog_period_in_ms
Description
The cycle when metadata log is periodically forced to be written to disk(in milliseconds). If force_mlog_period_in_ms = 0 it means force metadata log to be written to disk after each refreshment
Type
Int64
Default
100
Effective
After restarting system
tag_attribute_flush_interval
Name
tag_attribute_flush_interval
Description
interval num for tag and attribute records when force flushing to disk. When a certain amount of tag and attribute records is reached, they will be force flushed to disk. It is possible to lose at most tag_attribute_flush_interval records
Type
int32
Default
1000
Effective
Only allowed to be modified in first start up
tag_attribute_total_size
Name
tag_attribute_total_size
Description
The maximum persistence size of tags and attributes of each time series.
Type
int32
Default
700
Effective
Only allowed to be modified in first start up
schema_region_device_node_cache_size
Name
schema_region_device_node_cache_size
Description
The max num of device node, used for speeding up device query, cached in schemaRegion.
Type
Int32
Default
10000
Effective
After restarting system
max_measurement_num_of_internal_request
Name
max_measurement_num_of_internal_request
Description
When there's too many measurements in one create timeseries plan, the plan will be split to several sub plan, with measurement num no more than this param.
whether auto create the time series when a non-existed time series data comes
Type
true or false
Default
true
Effective
After restarting system
default_storage_group_level
Name
default_storage_group_level
Description
Database level when creating schema automatically is enabled. For example, if we receives a data point from root.sg0.d1.s2, we will set root.sg0 as the database if database level is 1. (root is level 0)
Type
integer
Default
1
Effective
After restarting system
boolean_string_infer_type
Name
boolean_string_infer_type
Description
To which type the values "true" and "false" should be reslved
Type
BOOLEAN or TEXT
Default
BOOLEAN
Effective
After restarting system
integer_string_infer_type
Name
integer_string_infer_type
Description
To which type an integer string like "67" in a query should be resolved
Type
INT32, INT64, DOUBLE, FLOAT or TEXT
Default
DOUBLE
Effective
After restarting system
floating_string_infer_type
Name
floating_string_infer_type
Description
To which type a floating number string like "6.7" in a query should be resolved
Type
DOUBLE, FLOAT or TEXT
Default
FLOAT
Effective
After restarting system
nan_string_infer_type
Name
nan_string_infer_type
Description
To which type the value NaN in a query should be resolved
Define the maximum age of data for which each tier is responsible
Type
long
Default
-1
Effective
After restarting system
max_waiting_time_when_insert_blocked
Name
max_waiting_time_when_insert_blocked
Description
When the waiting time(in ms) of an inserting exceeds this, throw an exception
Type
Int32
Default
10000
Effective
After restarting system
enable_discard_out_of_order_data
Name
enable_discard_out_of_order_data
Description
whether to discard out of order data
Type
Boolean
Default
false
Effective
After restarting system
handle_system_error
Name
handle_system_error
Description
What will the system do when unrecoverable error occurs
Type
String
Default
CHANGE_TO_READ_ONLY
Effective
After restarting system
memtable_size_threshold
Name
memtable_size_threshold
Description
max memtable size
Type
Long
Default
1073741824
Effective
when enable_mem_control is false & After restarting system
write_memory_variation_report_proportion
Name
write_memory_variation_report_proportion
Description
if memory cost of data region increased more than proportion of allocated memory for write, report to system
Type
Double
Default
0.001
Effective
After restarting system
enable_timed_flush_seq_memtable
Name
enable_timed_flush_seq_memtable
Description
whether to enable timed flush sequence memtable
Type
Boolean
Default
true
Effective
hot-load
seq_memtable_flush_interval_in_ms
Name
seq_memtable_flush_interval_in_ms
Description
if a memTable's created time is older than current time minus this, the memtable will be flushed to disk
Type
int32
Default
10800000
Effective
hot-load
seq_memtable_flush_check_interval_in_ms
Name
seq_memtable_flush_check_interval_in_ms
Description
the interval to check whether sequence memtables need flushing
Type
int32
Default
600000
Effective
hot-load
enable_timed_flush_unseq_memtable
Name
enable_timed_flush_unseq_memtable
Description
whether to enable timed flush unsequence memtable
Type
Boolean
Default
false
Effective
hot-load
unseq_memtable_flush_interval_in_ms
Name
unseq_memtable_flush_interval_in_ms
Description
if a memTable's created time is older than current time minus this, the memtable will be flushed to disk
Type
int32
Default
10800000
Effective
hot-load
unseq_memtable_flush_check_interval_in_ms
Name
unseq_memtable_flush_check_interval_in_ms
Description
the interval to check whether unsequence memtables need flushing
Type
int32
Default
600000
Effective
hot-load
tvlist_sort_algorithm
Name
tvlist_sort_algorithm
Description
the sort algorithm used in the memtable's TVList
Type
String
Default
TIM
Effective
After restarting system
avg_series_point_number_threshold
Name
avg_series_point_number_threshold
Description
max average number of point of each series in memtable
Type
int32
Default
100000
Effective
After restarting system
flush_thread_count
Name
flush_thread_count
Description
The thread number used to perform the operation when IoTDB writes data in memory to disk. If the value is less than or equal to 0, then the number of CPU cores installed on the machine is used. The default is 0.
Type
int32
Default
0
Effective
After restarting system
enable_partial_insert
Name
enable_partial_insert
Description
Whether continue to write other measurements if some measurements are failed in one insertion.
Type
Boolean
Default
true
Effective
After restarting system
recovery_log_interval_in_ms
Name
recovery_log_interval_in_ms
Description
the interval to log recover progress of each region when starting iotdb
Type
Int32
Default
5000
Effective
After restarting system
0.13_data_insert_adapt
Name
0.13_data_insert_adapt
Description
if using v0.13 client to insert data, set this configuration to true.
Type
Boolean
Default
false
Effective
After restarting system
device_path_cache_size
Name
device_path_cache_size
Description
The max size of the device path cache. This cache is for avoiding initialize duplicated device id object in write process
enable the compaction between sequence files and unsequence files
Type
Boolean
Default
true
Effective
hot-load
cross_selector
Name
cross_selector
Description
the task selector type of cross space compaction
Type
String
Default
rewrite
Effective
After restart system
cross_performer
Name
cross_performer
Description
the task performer type of cross space compaction. The options are read_point and fast, read_point is the default and fast is still under test
Type
String
Default
read_point
Effective
After restart system
inner_seq_selector
Name
inner_seq_selector
Description
the task selector type of inner sequence space compaction
Type
String
Default
size_tiered
Effective
After restart system
inner_seq_performer
Name
inner_seq_peformer
Description
the task performer type of inner sequence space compaction. The options are read_chunk and fast, read_chunk is the default and fast is still under test
Type
String
Default
read_chunk
Effective
After restart system
inner_unseq_selector
Name
inner_unseq_selector
Description
the task selector type of inner unsequence space compaction
Type
String
Default
size_tiered
Effective
After restart system
inner_unseq_performer
Name
inner_unseq_peformer
Description
the task performer type of inner unsequence space compaction. The options are read_point and fast, read_point is the default and fast is still under test
Type
String
Default
read_point
Effective
After restart system
compaction_priority
Name
compaction_priority
Description
Priority of compaction task. When it is BALANCE, system executes all types of compaction equally; when it is INNER_CROSS, system takes precedence over executing inner space compaction task; when it is CROSS_INNER, system takes precedence over executing cross space compaction task
Type
String
Default
BALANCE
Effective
After restart system
target_compaction_file_size
Name
target_compaction_file_size
Description
The target file size in compaction
Type
Int64
Default
2147483648
Effective
After restart system
target_chunk_size
Name
target_chunk_size
Description
The target size of compacted chunk
Type
Int64
Default
1048576
Effective
After restart system
target_chunk_point_num
Name
target_chunk_point_num
Description
The target point number of compacted chunk
Type
int32
Default
100000
Effective
After restart system
chunk_size_lower_bound_in_compaction
Name
chunk_size_lower_bound_in_compaction
Description
A source chunk will be deserialized in compaction when its size is less than this value
Type
Int64
Default
10240
Effective
After restart system
chunk_point_num_lower_bound_in_compaction
Name
chunk_point_num_lower_bound_in_compaction
Description
A source chunk will be deserialized in compaction when its point num is less than this value
Type
int32
Default
1000
Effective
After restart system
max_inner_compaction_candidate_file_num
Name
max_inner_compaction_candidate_file_num
Description
The max num of files encounter in inner space compaction
Type
int32
Default
30
Effective
After restart system
max_cross_compaction_file_num
Name
max_cross_compaction_candidate_file_num
Description
The max num of files encounter in cross space compaction
Type
int32
Default
500
Effective
After restart system
max_cross_compaction_file_size
Name
max_cross_compaction_candidate_file_size
Description
The max size of files encounter in cross space compaction
Type
Int64
Default
5368709120
Effective
After restart system
cross_compaction_file_selection_time_budget
Name
cross_compaction_file_selection_time_budget
Description
Time budget for cross space compaction file selection
Type
int32
Default
30000
Effective
After restart system
compaction_thread_count
Name
compaction_thread_count
Description
thread num to execute compaction
Type
int32
Default
10
Effective
hot-load
compaction_schedule_interval_in_ms
Name
compaction_schedule_interval_in_ms
Description
interval of scheduling compaction
Type
Int64
Default
60000
Effective
After restart system
compaction_submission_interval_in_ms
Name
compaction_submission_interval_in_ms
Description
interval of submitting compaction task
Type
Int64
Default
60000
Effective
After restart system
compaction_write_throughput_mb_per_sec
Name
compaction_write_throughput_mb_per_sec
Description
The write rate of all compaction tasks in MB/s
Type
int32
Default
16
Effective
After restart system
sub_compaction_thread_count
Name
sub_compaction_thread_count
Description
the number of sub-compaction threads to accelerate cross space compaction
Type
Int32
Default
4
Effective
hot-load
compaction_validation_level
名字
compaction_validation_level
Description
The level of validation after compaction. NONE: the validation after compaction is disabled. RESOURCE_ONLY: the validation after compaction check tsfile resource only. RESOURCE_AND_TSFILE: the validation after compaction check resource and file.
The write mode of wal. For DISABLE mode, the system will disable wal. For SYNC mode, the system will submit wal synchronously, write request will not return until its wal is fsynced to the disk successfully. For ASYNC mode, the system will submit wal asynchronously, write request will return immediately no matter its wal is fsynced to the disk successfully.
Type
String
Default
ASYNC
Effective
After restart system
max_wal_nodes_num
Name
max_wal_nodes_num
Description
Max number of wal nodes, each node corresponds to one wal directory. The default value 0 means the number is determined by the system.
Type
int32
Default
0
Effective
After restart system
wal_async_mode_fsync_delay_in_ms
Name
wal_async_mode_fsync_delay_in_ms
Description
Duration a wal flush operation will wait before calling fsync in the async mode
Type
int32
Default
1000
Effective
hot-load
wal_sync_mode_fsync_delay_in_ms
Name
wal_sync_mode_fsync_delay_in_ms
Description
Duration a wal flush operation will wait before calling fsync in the sync mode
Type
int32
Default
3
Effective
hot-load
wal_buffer_size_in_byte
Name
wal_buffer_size_in_byte
Description
Buffer size of each wal node
Type
int32
Default
33554432
Effective
After restart system
wal_buffer_queue_capacity
Name
wal_buffer_queue_capacity
Description
Blocking queue capacity of each wal buffer
Type
int32
Default
500
Effective
After restart system
wal_file_size_threshold_in_byte
Name
wal_file_size_threshold_in_byte
Description
Size threshold of each wal file
Type
int32
Default
31457280
Effective
hot-load
wal_min_effective_info_ratio
Name
wal_min_effective_info_ratio
Description
Minimum ratio of effective information in wal files
Type
double
Default
0.1
Effective
hot-load
wal_memtable_snapshot_threshold_in_byte
Name
wal_memtable_snapshot_threshold_in_byte
Description
MemTable size threshold for triggering MemTable snapshot in wal
Type
int64
Default
8388608
Effective
hot-load
max_wal_memtable_snapshot_num
Name
max_wal_memtable_snapshot_num
Description
MemTable's max snapshot number in wal
Type
int32
Default
1
Effective
hot-load
delete_wal_files_period_in_ms
Name
delete_wal_files_period_in_ms
Description
The period when outdated wal files are periodically deleted
The maximum size of a single page written in memory when each column in memory is written (in bytes)
Type
int32
Default
65536
Effective
hot-load
max_number_of_points_in_page
Name
max_number_of_points_in_page
Description
The maximum number of data points (timestamps - valued groups) contained in a page
Type
int32
Default
10000
Effective
hot-load
pattern_matching_threshold
Name
pattern_matching_threshold
Description
Max matching time of regex pattern
Type
int32
Default
1000000
Effective
hot-load
max_degree_of_index_node
Name
max_degree_of_index_node
Description
The maximum degree of the metadata index tree (that is, the max number of each node's children)
Type
int32
Default
256
Effective
Only allowed to be modified in first start up
max_string_length
Name
max_string_length
Description
The maximum length of a single string (number of character)
Type
int32
Default
128
Effective
hot-load
value_encoder
Name
value_encoder
Description
Encoding type of value column
Type
Enum String: “TS_2DIFF”,“PLAIN”,“RLE”
Default
PLAIN
Effective
hot-load
float_precision
Name
float_precision
Description
The precision of the floating point number.(The number of digits after the decimal point)
Type
int32
Default
The default is 2 digits. Note: The 32-bit floating point number has a decimal precision of 7 bits, and the 64-bit floating point number has a decimal precision of 15 bits. If the setting is out of the range, it will have no practical significance.
Effective
hot-load
compressor
Name
compressor
Description
Data compression method; Time compression method in aligned timeseries
The false positive rate of bloom filter in each TsFile. Bloom filter checks whether a given time series is in the tsfile before loading metadata. This can improve the performance of loading metadata and skip the tsfile that doesn't contain specified time series. If you want to learn more about its mechanism, you can refer to: wiki page of bloom filter.
Used to estimate the memory usage of text fields in a UDF query. It is recommended to set this value to be slightly larger than the average length of all texts.
Type
int32
Default
48
Effective
After restarting system
udf_memory_budget_in_mb
Name
udf_memory_budget_in_mb
Description
How much memory may be used in ONE UDF query (in MB). The upper limit is 20% of allocated memory for read.
Set the white list of IP addresses of the sender of the synchronization, which is expressed in the form of network segments, and multiple network segments are separated by commas. When the sender synchronizes data to the receiver, the receiver allows synchronization only when the IP address of the sender is within the network segment set in the white list. If the whitelist is empty, the receiver does not allow any sender to synchronize data. By default, the receiver rejects the synchronization request of all IP addresses except 127.0.0.1. When configuring this parameter, please ensure that all DataNode addresses on the sender are set.
Type
String
Default
127.0.0.1/32
Effective
hot-load
max_number_of_sync_file_retry
Name
max_number_of_sync_file_retry
Description
The maximum number of retries when the sender fails to synchronize files to the receiver.
