NasaLogSparkStreaming - Details for Query 10

Details for Query 10

Submitted Time: 2024/12/11 23:57:44
Duration: 13 s
Succeeded Jobs: 6 7

Show the Stage ID and Task ID that corresponds to the max metric

digraph G { 0 [labelType="html" label=" WriteToDataSourceV2 "]; subgraph cluster1 { isCluster="true"; label="WholeStageCodegen (7)\n \nduration: total (min, med, max (stageId: taskId))\n8 ms (0 ms, 0 ms, 3 ms (stage 19.0: task 1623))"; 2 [labelType="html" label="Sort sort time total (min, med, max (stageId: taskId)) 0 ms (0 ms, 0 ms, 0 ms (stage 19.0: task 1622)) peak memory total (min, med, max (stageId: taskId)) 512.6 MiB (64.0 KiB, 64.1 MiB, 64.1 MiB (stage 19.0: task 1622)) spill size total (min, med, max (stageId: taskId)) 0.0 B (0.0 B, 0.0 B, 0.0 B (stage 19.0: task 1622))"]; } 3 [labelType="html" label="Exchange shuffle records written: 12 shuffle write time total (min, med, max (stageId: taskId)) 6 ms (0 ms, 0 ms, 1 ms (stage 18.0: task 1472)) records read: 12 local bytes read total (min, med, max (stageId: taskId)) 942.0 B (0.0 B, 107.0 B, 172.0 B (stage 19.0: task 1628)) fetch wait time total (min, med, max (stageId: taskId)) 1 ms (0 ms, 0 ms, 1 ms (stage 19.0: task 1622)) remote bytes read total (min, med, max (stageId: taskId)) 281.0 B (0.0 B, 0.0 B, 174.0 B (stage 19.0: task 1628)) local blocks read: 9 remote blocks read: 3 data size total (min, med, max (stageId: taskId)) 840.0 B (0.0 B, 0.0 B, 80.0 B (stage 18.0: task 1522)) shuffle bytes written total (min, med, max (stageId: taskId)) 1223.0 B (0.0 B, 0.0 B, 115.0 B (stage 18.0: task 1522))"]; subgraph cluster4 { isCluster="true"; label="WholeStageCodegen (6)\n \nduration: total (min, med, max (stageId: taskId))\n450 ms (0 ms, 0 ms, 7 ms (stage 18.0: task 1427))"; 5 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 6 ms (0 ms, 0 ms, 2 ms (stage 16.0: task 1408)) peak memory total (min, med, max (stageId: taskId)) 1636.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 18.0: task 1439)) number of output rows: 24 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 18.0: task 1439))"]; } 6 [labelType="html" label="StateStoreSave number of total state rows: 24 memory used by state total (min, med, max (stageId: taskId)) 129.1 KiB (232.0 B, 400.0 B, 736.0 B (stage 16.0: task 1325)) count of cache hit on states cache in provider: 4,400 number of output rows: 24 estimated size of state only on current version total (min, med, max (stageId: taskId)) 39.6 KiB (88.0 B, 88.0 B, 320.0 B (stage 18.0: task 1522)) time to commit changes total (min, med, max (stageId: taskId)) 16.6 s (22 ms, 28 ms, 108 ms (stage 18.0: task 1545)) time to remove total (min, med, max (stageId: taskId)) 0 ms (0 ms, 0 ms, 0 ms (stage 18.0: task 1422)) number of updated state rows: 8 time to update total (min, med, max (stageId: taskId)) 256 ms (0 ms, 0 ms, 69 ms (stage 16.0: task 1408))"]; subgraph cluster7 { isCluster="true"; label="WholeStageCodegen (5)\n \nduration: total (min, med, max (stageId: taskId))\n257 ms (0 ms, 0 ms, 69 ms (stage 16.0: task 1408))"; 8 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 45 ms (0 ms, 0 ms, 45 ms (stage 16.0: task 1408)) peak memory total (min, med, max (stageId: taskId)) 612.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 18.0: task 1522)) number of output rows: 8 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 18.0: task 1522))"]; } 9 [labelType="html" label="StateStoreRestore number of output rows: 8"]; subgraph cluster10 { isCluster="true"; label="WholeStageCodegen (4)\n \nduration: total (min, med, max (stageId: taskId))\n936 ms (0 ms, 1 ms, 52 ms (stage 16.0: task 1408))"; 11 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 18 ms (0 ms, 0 ms, 3 ms (stage 18.0: task 1522)) peak memory total (min, med, max (stageId: taskId)) 612.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 18.0: task 1522)) number of output rows: 8 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 18.0: task 1522))"]; } 12 [labelType="html" label="Exchange shuffle records written: 8 shuffle write time total (min, med, max (stageId: taskId)) 7 ms (2 ms, 4 ms, 4 ms (stage 15.0: task 1220)) records read: 16 local bytes read total (min, med, max (stageId: taskId)) 884.0 B (0.0 B, 0.0 B, 114.0 B (stage 18.0: task 1522)) fetch wait time total (min, med, max (stageId: taskId)) 8 ms (0 ms, 0 ms, 2 ms (stage 18.0: task 1522)) remote bytes read total (min, med, max (stageId: taskId)) 884.0 B (0.0 B, 0.0 B, 114.0 B (stage 18.0: task 1522)) local blocks read: 8 remote blocks read: 8 data size total (min, med, max (stageId: taskId)) 608.0 B (304.0 B, 304.0 B, 304.0 B (stage 15.0: task 1221)) shuffle bytes written total (min, med, max (stageId: taskId)) 884.0 B (442.0 B, 442.0 B, 442.0 B (stage 15.0: task 1221))"]; subgraph cluster13 { isCluster="true"; label="WholeStageCodegen (3)\n \nduration: total (min, med, max (stageId: taskId))\n16 ms (5 ms, 11 ms, 11 ms (stage 15.0: task 1220))"; 14 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 3 ms (1 ms, 2 ms, 2 ms (stage 15.0: task 1220)) peak memory total (min, med, max (stageId: taskId)) 128.5 MiB (64.3 MiB, 64.3 MiB, 64.3 MiB (stage 15.0: task 1221)) number of output rows: 8 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 15.0: task 1221))"]; 15 [labelType="html" label=" Project "]; 16 [labelType="html" label="Filter number of output rows: 28"]; 17 [labelType="html" label="Expand number of output rows: 28"]; 18 [labelType="html" label=" Project "]; } 19 [labelType="html" label=" EventTimeWatermark "]; subgraph cluster20 { isCluster="true"; label="WholeStageCodegen (2)\n \nduration: total (min, med, max (stageId: taskId))\n9 ms (2 ms, 7 ms, 7 ms (stage 15.0: task 1220))"; 21 [labelType="html" label=" Project "]; } 22 [labelType="html" label="Generate number of output rows: 14"]; subgraph cluster23 { isCluster="true"; label="WholeStageCodegen (1)\n \nduration: total (min, med, max (stageId: taskId))\n13 ms (3 ms, 10 ms, 10 ms (stage 15.0: task 1220))"; 24 [labelType="html" label=" Project "]; } 25 [labelType="html" label="MicroBatchScan number of output rows: 14"]; 2->0; 3->2; 5->3; 6->5; 8->6; 9->8; 11->9; 12->11; 14->12; 15->14; 16->15; 17->16; 18->17; 19->18; 21->19; 22->21; 24->22; 25->24; }

WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@341b884c

Sort [total_bytes#68 DESC NULLS LAST], true, 0

WholeStageCodegen (7)

Exchange rangepartitioning(total_bytes#68 DESC NULLS LAST, 200), true, [id=#436]

HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[sum(cast(bytes#47 as double))])

WholeStageCodegen (6)

StateStoreSave [window#69-T3000ms, hostname#12], state info [ checkpoint = file:/tmp/spark-checkpoints/state, runId = e0c08a0a-5cf3-468b-888d-995173b23fbd, opId = 0, ver = 3, numPartitions = 200], Complete, 1733961447433, 2

HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[merge_sum(cast(bytes#47 as double))])

WholeStageCodegen (5)

StateStoreRestore [window#69-T3000ms, hostname#12], state info [ checkpoint = file:/tmp/spark-checkpoints/state, runId = e0c08a0a-5cf3-468b-888d-995173b23fbd, opId = 0, ver = 3, numPartitions = 200], 2

HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[merge_sum(cast(bytes#47 as double))])

WholeStageCodegen (4)

Exchange hashpartitioning(window#69-T3000ms, hostname#12, 200), true, [id=#424]

HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[partial_sum(cast(bytes#47 as double))])

Project [window#69-T3000ms, hostname#12, bytes#47]

Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#69-T3000ms)) AND (timestamp#1-T3000ms >= window#69-T3000ms.start)) AND (timestamp#1-T3000ms < window#69-T3000ms.end))

Expand [ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 0) - 2) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 0) - 2) * 30000000) + 60000000), LongType, TimestampType)), timestamp#1-T3000ms, hostname#12, bytes#47), ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 1) - 2) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 1) - 2) * 30000000) + 60000000), LongType, TimestampType)), timestamp#1-T3000ms, hostname#12, bytes#47)], [window#69-T3000ms, timestamp#1-T3000ms, hostname#12, bytes#47]

Project [timestamp#1-T3000ms, split(logs#5, ,, -1)[1] AS hostname#12, split(logs#5, ,, -1)[6] AS bytes#47]

WholeStageCodegen (3)

EventTimeWatermark timestamp#1: timestamp, 3 seconds

Project [logs#5, timestamp#1]

WholeStageCodegen (2)

Generate explode(split(value#0, , -1)), [timestamp#1], false, [logs#5]

Project [value#0, timestamp#1]

WholeStageCodegen (1)

MicroBatchScan[value#0, timestamp#1] class org.apache.spark.sql.execution.streaming.sources.TextSocketTable$$anon$1

Details

== Parsed Logical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@341b884c
+- Sort [total_bytes#68 DESC NULLS LAST], true
   +- Aggregate [window#69-T3000ms, hostname#12], [window#69-T3000ms AS window#57-T3000ms, hostname#12, sum(cast(bytes#47 as double)) AS total_bytes#68]
      +- Filter ((timestamp#1-T3000ms >= window#69-T3000ms.start) AND (timestamp#1-T3000ms < window#69-T3000ms.end))
         +- Expand [ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(0 as bigint)) - cast(2 as bigint)) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion((((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(0 as bigint)) - cast(2 as bigint)) * 30000000) + 0) + 60000000), LongType, TimestampType)), logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, bytes#47), ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(1 as bigint)) - cast(2 as bigint)) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion((((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(1 as bigint)) - cast(2 as bigint)) * 30000000) + 0) + 60000000), LongType, TimestampType)), logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, bytes#47)], [window#69-T3000ms, logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, bytes#47]
            +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, split(logs#5, ,, -1)[6] AS bytes#47]
               +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, split(logs#5, ,, -1)[5] AS responsecode#38]
                  +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, split(logs#5, ,, -1)[4] AS url#30]
                     +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, split(logs#5, ,, -1)[3] AS method#23]
                        +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, split(logs#5, ,, -1)[2] AS time#17]
                           +- Project [logs#5, timestamp#1-T3000ms, idx#8, split(logs#5, ,, -1)[1] AS hostname#12]
                              +- Project [logs#5, timestamp#1-T3000ms, split(logs#5, ,, -1)[0] AS idx#8]
                                 +- EventTimeWatermark timestamp#1: timestamp, 3 seconds
                                    +- Project [logs#5, timestamp#1]
                                       +- Generate explode(split(value#0,  , -1)), false, [logs#5]
                                          +- StreamingDataSourceV2Relation [value#0, timestamp#1], org.apache.spark.sql.execution.streaming.sources.TextSocketTable$$anon$1@64e6357e, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 32, 46

== Analyzed Logical Plan ==

WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@341b884c
+- Sort [total_bytes#68 DESC NULLS LAST], true
   +- Aggregate [window#69-T3000ms, hostname#12], [window#69-T3000ms AS window#57-T3000ms, hostname#12, sum(cast(bytes#47 as double)) AS total_bytes#68]
      +- Filter ((timestamp#1-T3000ms >= window#69-T3000ms.start) AND (timestamp#1-T3000ms < window#69-T3000ms.end))
         +- Expand [ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(0 as bigint)) - cast(2 as bigint)) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion((((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(0 as bigint)) - cast(2 as bigint)) * 30000000) + 0) + 60000000), LongType, TimestampType)), logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, bytes#47), ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(1 as bigint)) - cast(2 as bigint)) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion((((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) + cast(1 as bigint)) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / cast(30000000 as double))) END + cast(1 as bigint)) - cast(2 as bigint)) * 30000000) + 0) + 60000000), LongType, TimestampType)), logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, bytes#47)], [window#69-T3000ms, logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, bytes#47]
            +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, responsecode#38, split(logs#5, ,, -1)[6] AS bytes#47]
               +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, url#30, split(logs#5, ,, -1)[5] AS responsecode#38]
                  +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, method#23, split(logs#5, ,, -1)[4] AS url#30]
                     +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, time#17, split(logs#5, ,, -1)[3] AS method#23]
                        +- Project [logs#5, timestamp#1-T3000ms, idx#8, hostname#12, split(logs#5, ,, -1)[2] AS time#17]
                           +- Project [logs#5, timestamp#1-T3000ms, idx#8, split(logs#5, ,, -1)[1] AS hostname#12]
                              +- Project [logs#5, timestamp#1-T3000ms, split(logs#5, ,, -1)[0] AS idx#8]
                                 +- EventTimeWatermark timestamp#1: timestamp, 3 seconds
                                    +- Project [logs#5, timestamp#1]
                                       +- Generate explode(split(value#0,  , -1)), false, [logs#5]
                                          +- StreamingDataSourceV2Relation [value#0, timestamp#1], org.apache.spark.sql.execution.streaming.sources.TextSocketTable$$anon$1@64e6357e, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 32, 46

== Optimized Logical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@341b884c
+- Sort [total_bytes#68 DESC NULLS LAST], true
   +- Aggregate [window#69-T3000ms, hostname#12], [window#69-T3000ms AS window#57-T3000ms, hostname#12, sum(cast(bytes#47 as double)) AS total_bytes#68]
      +- Project [window#69-T3000ms, hostname#12, bytes#47]
         +- Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#69-T3000ms)) AND (timestamp#1-T3000ms >= window#69-T3000ms.start)) AND (timestamp#1-T3000ms < window#69-T3000ms.end))
            +- Expand [ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 0) - 2) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 0) - 2) * 30000000) + 60000000), LongType, TimestampType)), timestamp#1-T3000ms, hostname#12, bytes#47), ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 1) - 2) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 1) - 2) * 30000000) + 60000000), LongType, TimestampType)), timestamp#1-T3000ms, hostname#12, bytes#47)], [window#69-T3000ms, timestamp#1-T3000ms, hostname#12, bytes#47]
               +- Project [timestamp#1-T3000ms, split(logs#5, ,, -1)[1] AS hostname#12, split(logs#5, ,, -1)[6] AS bytes#47]
                  +- EventTimeWatermark timestamp#1: timestamp, 3 seconds
                     +- Project [logs#5, timestamp#1]
                        +- Generate explode(split(value#0,  , -1)), [0], false, [logs#5]
                           +- StreamingDataSourceV2Relation [value#0, timestamp#1], org.apache.spark.sql.execution.streaming.sources.TextSocketTable$$anon$1@64e6357e, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 32, 46

== Physical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@341b884c
+- *(7) Sort [total_bytes#68 DESC NULLS LAST], true, 0
   +- Exchange rangepartitioning(total_bytes#68 DESC NULLS LAST, 200), true, [id=#436]
      +- *(6) HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[sum(cast(bytes#47 as double))], output=[window#57-T3000ms, hostname#12, total_bytes#68])
         +- StateStoreSave [window#69-T3000ms, hostname#12], state info [ checkpoint = file:/tmp/spark-checkpoints/state, runId = e0c08a0a-5cf3-468b-888d-995173b23fbd, opId = 0, ver = 3, numPartitions = 200], Complete, 1733961447433, 2
            +- *(5) HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[merge_sum(cast(bytes#47 as double))], output=[window#69-T3000ms, hostname#12, sum#77])
               +- StateStoreRestore [window#69-T3000ms, hostname#12], state info [ checkpoint = file:/tmp/spark-checkpoints/state, runId = e0c08a0a-5cf3-468b-888d-995173b23fbd, opId = 0, ver = 3, numPartitions = 200], 2
                  +- *(4) HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[merge_sum(cast(bytes#47 as double))], output=[window#69-T3000ms, hostname#12, sum#77])
                     +- Exchange hashpartitioning(window#69-T3000ms, hostname#12, 200), true, [id=#424]
                        +- *(3) HashAggregate(keys=[window#69-T3000ms, hostname#12], functions=[partial_sum(cast(bytes#47 as double))], output=[window#69-T3000ms, hostname#12, sum#77])
                           +- *(3) Project [window#69-T3000ms, hostname#12, bytes#47]
                              +- *(3) Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#69-T3000ms)) AND (timestamp#1-T3000ms >= window#69-T3000ms.start)) AND (timestamp#1-T3000ms < window#69-T3000ms.end))
                                 +- *(3) Expand [ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 0) - 2) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 0) - 2) * 30000000) + 60000000), LongType, TimestampType)), timestamp#1-T3000ms, hostname#12, bytes#47), ArrayBuffer(named_struct(start, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 1) - 2) * 30000000) + 0), LongType, TimestampType), end, precisetimestampconversion(((((CASE WHEN (cast(CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) as double) = (cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) THEN (CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) + 1) ELSE CEIL((cast((precisetimestampconversion(timestamp#1-T3000ms, TimestampType, LongType) - 0) as double) / 3.0E7)) END + 1) - 2) * 30000000) + 60000000), LongType, TimestampType)), timestamp#1-T3000ms, hostname#12, bytes#47)], [window#69-T3000ms, timestamp#1-T3000ms, hostname#12, bytes#47]
                                    +- *(3) Project [timestamp#1-T3000ms, split(logs#5, ,, -1)[1] AS hostname#12, split(logs#5, ,, -1)[6] AS bytes#47]
                                       +- EventTimeWatermark timestamp#1: timestamp, 3 seconds
                                          +- *(2) Project [logs#5, timestamp#1]
                                             +- Generate explode(split(value#0,  , -1)), [timestamp#1], false, [logs#5]
                                                +- *(1) Project [value#0, timestamp#1]
                                                   +- MicroBatchScan[value#0, timestamp#1] class org.apache.spark.sql.execution.streaming.sources.TextSocketTable$$anon$1