[GitHub] [iotdb] HeimingZ commented on a change in pull request #5320: [IOTDB-1614] New WAL

[GitBox <gi...@apache.org>]

HeimingZ commented on a change in pull request #5320:
URL: https://github.com/apache/iotdb/pull/5320#discussion_r834160450



##########
File path: server/src/assembly/resources/conf/iotdb-engine.properties
##########
@@ -50,25 +50,45 @@ rpc_port=6667
 ### Write Ahead Log Configuration
 ####################
 
-# Is insert ahead log enable
-# Datatype: boolean
-# enable_wal=true
+# Write mode of wal
+# The details of these three modes are as follows:
+# 1. DISABLE: the system will disable wal.
+# 2. SYNC: the system will submit wal synchronously, write request will not return until its wal is fsynced to the disk successfully.
+# 3. ASYNC: the system will submit wal asynchronously, write request will return immediately no matter its wal is fsynced to the disk successfully.
+# The write performance order is DISABLE > ASYNC > SYNC, but only SYNC mode can ensure data durability.
+# wal_mode=SYNC
+
+# Duration a wal flush operation will wait before calling fsync
+# A duration greater than 0 batches multiple wal fsync calls into one. This is useful when disks are slow or WAL write contention exists.
+# Datatype: long
+# sync_wal_delay_in_ms=0
 
-# Add a switch to drop ouf-of-order data
-# Out-of-order data will impact the aggregation query a lot. Users may not care about discarding some out-of-order data.
-# Datatype: boolean
-# enable_discard_out_of_order_data=false
+# Max number of wal nodes, each node corresponds to one wal directory
+# The default value 0 means the concurrent wal number will be 2 * 'number of wal dirs'.
+# Datatype: int
+# max_wal_num=0
 
-# When a certain amount of insert ahead log is reached, it will be flushed to disk
-# It is possible to lose at most flush_wal_threshold operations
+# Buffer size of each wal node
+# If it sets a value smaller than 0, use the default value 16777216 bytes (16MB).
 # Datatype: int
-# flush_wal_threshold=10000
+# wal_buffer_size_in_byte=16777216
 
-# The cycle when insert ahead log is periodically forced to be written to disk(in milliseconds)
-# If force_wal_period_in_ms = 0 it means force insert ahead log to be written to disk after each refreshment
-# Set this parameter to 0 may slow down the ingestion on slow disk.
-# Datatype: long
-# force_wal_period_in_ms=100
+# Buffer entry size of each wal buffer
+# If it sets a value smaller than 0, use the default value 16384 bytes (16KB).
+# Datatype: int
+# wal_buffer_entry_size_in_byte=16384
+
+# Max storage space for wal
+# The default value 0 means the storage space will not be controlled.
+# Notice: If this parameter is too small, the write performance may decline.
+# Datatype: int
+# wal_storage_space_in_mb=0

Review comment:
       Default 0 means the storage space will controlled by iotdb itself, you can see [delete method](https://github.com/apache/iotdb/pull/5320/files#diff-672f4b0027b7b2d89495f695447243e85d264fc61a37ee2dcbbcf01a3f59cf0aR166-R181) of WALManager. Besides, related functions of this parameter is still under development.




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: reviews-unsubscribe@iotdb.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org