Hey Everyone, I am loading data in druid here is spec file { "type": "index_parallel", "spec": { "ioConfig": { "type": "index_parallel", "inputSource": { "type": "local", "baseDir": "/home/hadoop/ipdr", "filter": "1709521742_ipdr_3" }, "inputFormat": { "type": "tsv", "findColumnsFromHeader": false, "delimiter": "|", "skipHeaderRows": 0, "columns": [ "Source_ip", "Source_port", "Destination_ip", "Destination_port", "protocol_l3", "protocol_l4", "Protocol_l7", "Application", "Category", "calling_number", "called_number", "Packets_sent", "Bytes_sent", "Packets_received", "Bytes_received", "Start_time", "End_time", "subscriber_id", "operator_id", "area_id" ] }, "appendToExisting": false }, "tuningConfig": { "type": "index_parallel", "partitionsSpec": { "type": "dynamic" }, "maxNumConcurrentSubTasks": 20 }, "dataSchema": { "dataSource": "header", "timestampSpec": { "column": "End_time", "format": "posix" }, "transformSpec": {}, "dimensionsSpec": { "useSchemaDiscovery": true, "dimensionExclusions": [] }, "granularitySpec": { "queryGranularity": "hour", "rollup": false, "segmentGranularity": "hour" } } } } the file size is of neary 2gb, It is taking about 10 to 15 mins to load a file, how I can make it fast to load a file

Is it only one file? If so, then you may need to split it up into multiple files.

Basically I receive 2 to 3 files in one minute

If you split the files up, you can you increased number of workers. Single file cannot be split into different workers.

Okay i will try it