NasaLogSparkStreaming - Details for Query 72

Details for Query 72

Submitted Time: 2024/12/04 20:19:32
Duration: 15 s
Succeeded Jobs: 26 27

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))\n13 ms (0 ms, 1 ms, 11 ms (stage 38.0: task 1653))"; 2 [labelType="html" label="Sort sort time total (min, med, max (stageId: taskId)) 0 ms (0 ms, 0 ms, 0 ms (stage 38.0: task 1654)) peak memory total (min, med, max (stageId: taskId)) 192.3 MiB (64.0 KiB, 64.1 MiB, 64.1 MiB (stage 38.0: task 1654)) spill size total (min, med, max (stageId: taskId)) 0.0 B (0.0 B, 0.0 B, 0.0 B (stage 38.0: task 1654))"]; } 3 [labelType="html" label="Exchange shuffle records written: 8 shuffle write time total (min, med, max (stageId: taskId)) 4 ms (0 ms, 0 ms, 1 ms (stage 37.0: task 1460)) records read: 8 local bytes read total (min, med, max (stageId: taskId)) 629.0 B (0.0 B, 214.0 B, 230.0 B (stage 38.0: task 1655)) fetch wait time total (min, med, max (stageId: taskId)) 7 ms (0 ms, 0 ms, 7 ms (stage 38.0: task 1653)) remote bytes read total (min, med, max (stageId: taskId)) 161.0 B (0.0 B, 0.0 B, 161.0 B (stage 38.0: task 1653)) local blocks read: 6 remote blocks read: 2 data size total (min, med, max (stageId: taskId)) 544.0 B (0.0 B, 0.0 B, 80.0 B (stage 37.0: task 1530)) shuffle bytes written total (min, med, max (stageId: taskId)) 790.0 B (0.0 B, 0.0 B, 115.0 B (stage 37.0: task 1530))"]; subgraph cluster4 { isCluster="true"; label="WholeStageCodegen (6)\n \nduration: total (min, med, max (stageId: taskId))\n778 ms (0 ms, 1 ms, 91 ms (stage 37.0: task 1556))"; 5 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 130 ms (0 ms, 0 ms, 86 ms (stage 37.0: task 1556)) peak memory total (min, med, max (stageId: taskId)) 1124.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 37.0: task 1460)) number of output rows: 16 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 37.0: task 1460))"]; } 6 [labelType="html" label="StateStoreSave number of total state rows: 16 memory used by state total (min, med, max (stageId: taskId)) 127.2 KiB (232.0 B, 400.0 B, 728.0 B (stage 35.0: task 1311)) count of cache hit on states cache in provider: 2,800 number of output rows: 16 estimated size of state only on current version total (min, med, max (stageId: taskId)) 37.8 KiB (88.0 B, 88.0 B, 320.0 B (stage 37.0: task 1530)) time to commit changes total (min, med, max (stageId: taskId)) 19.7 s (27 ms, 37 ms, 101 ms (stage 37.0: task 1585)) time to remove total (min, med, max (stageId: taskId)) 0 ms (0 ms, 0 ms, 0 ms (stage 37.0: task 1454)) number of updated state rows: 8 time to update total (min, med, max (stageId: taskId)) 239 ms (0 ms, 0 ms, 13 ms (stage 37.0: task 1607))"]; subgraph cluster7 { isCluster="true"; label="WholeStageCodegen (5)\n \nduration: total (min, med, max (stageId: taskId))\n243 ms (0 ms, 0 ms, 13 ms (stage 37.0: task 1607))"; 8 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 0 ms (0 ms, 0 ms, 0 ms (stage 37.0: task 1454)) peak memory total (min, med, max (stageId: taskId)) 612.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 37.0: task 1511)) number of output rows: 8 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 37.0: task 1511))"]; } 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))\n1.0 s (1 ms, 2 ms, 12 ms (stage 35.0: task 1373))"; 11 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 19 ms (0 ms, 0 ms, 4 ms (stage 35.0: task 1311)) peak memory total (min, med, max (stageId: taskId)) 612.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 37.0: task 1511)) number of output rows: 8 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 37.0: task 1511))"]; } 12 [labelType="html" label="Exchange shuffle records written: 8 shuffle write time total (min, med, max (stageId: taskId)) 8 ms (4 ms, 4 ms, 4 ms (stage 34.0: task 1251)) records read: 16 local bytes read total (min, med, max (stageId: taskId)) 888.0 B (0.0 B, 0.0 B, 115.0 B (stage 37.0: task 1530)) fetch wait time total (min, med, max (stageId: taskId)) 13 ms (0 ms, 0 ms, 3 ms (stage 35.0: task 1311)) remote bytes read total (min, med, max (stageId: taskId)) 888.0 B (0.0 B, 0.0 B, 115.0 B (stage 37.0: task 1530)) 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 34.0: task 1252)) shuffle bytes written total (min, med, max (stageId: taskId)) 888.0 B (444.0 B, 444.0 B, 444.0 B (stage 34.0: task 1252))"]; subgraph cluster13 { isCluster="true"; label="WholeStageCodegen (3)\n \nduration: total (min, med, max (stageId: taskId))\n16 ms (6 ms, 10 ms, 10 ms (stage 34.0: task 1251))"; 14 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 4 ms (1 ms, 3 ms, 3 ms (stage 34.0: task 1251)) peak memory total (min, med, max (stageId: taskId)) 128.5 MiB (64.3 MiB, 64.3 MiB, 64.3 MiB (stage 34.0: task 1252)) number of output rows: 8 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 34.0: task 1252))"]; 15 [labelType="html" label=" Project "]; 16 [labelType="html" label="Filter number of output rows: 32"]; 17 [labelType="html" label="Expand number of output rows: 32"]; 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 (3 ms, 6 ms, 6 ms (stage 34.0: task 1251))"; 21 [labelType="html" label=" Project "]; } 22 [labelType="html" label="Generate number of output rows: 16"]; subgraph cluster23 { isCluster="true"; label="WholeStageCodegen (1)\n \nduration: total (min, med, max (stageId: taskId))\n12 ms (4 ms, 8 ms, 8 ms (stage 34.0: task 1251))"; 24 [labelType="html" label=" Project "]; } 25 [labelType="html" label="MicroBatchScan number of output rows: 16"]; 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@702a5285

Sort [total_bytes#858 DESC NULLS LAST], true, 0

WholeStageCodegen (7)

Exchange rangepartitioning(total_bytes#858 DESC NULLS LAST, 200), true, [id=#1212]

HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[sum(cast(bytes#175 as double))])

WholeStageCodegen (6)

StateStoreSave [window#859-T3000ms, hostname#140], state info [ checkpoint = file:/tmp/temporary-fcb72abb-9143-4d38-a65d-25baf17e34c7/state, runId = 49431645-20e3-4737-9dcc-8f334da22f1d, opId = 0, ver = 2, numPartitions = 200], Complete, 1733343553054, 2

HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[merge_sum(cast(bytes#175 as double))])

WholeStageCodegen (5)

StateStoreRestore [window#859-T3000ms, hostname#140], state info [ checkpoint = file:/tmp/temporary-fcb72abb-9143-4d38-a65d-25baf17e34c7/state, runId = 49431645-20e3-4737-9dcc-8f334da22f1d, opId = 0, ver = 2, numPartitions = 200], 2

HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[merge_sum(cast(bytes#175 as double))])

WholeStageCodegen (4)

Exchange hashpartitioning(window#859-T3000ms, hostname#140, 200), true, [id=#1200]

HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[partial_sum(cast(bytes#175 as double))])

Project [window#859-T3000ms, hostname#140, bytes#175]

Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#859-T3000ms)) AND (timestamp#1-T3000ms >= window#859-T3000ms.start)) AND (timestamp#1-T3000ms < window#859-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#140, bytes#175), 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#140, bytes#175)], [window#859-T3000ms, timestamp#1-T3000ms, hostname#140, bytes#175]

Project [timestamp#1-T3000ms, split(logs#5, ,, -1)[1] AS hostname#140, split(logs#5, ,, -1)[6] AS bytes#175]

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@702a5285
+- Sort [total_bytes#858 DESC NULLS LAST], true
   +- Aggregate [window#859-T3000ms, hostname#140], [window#859-T3000ms AS window#847-T3000ms, hostname#140, sum(cast(bytes#175 as double)) AS total_bytes#858]
      +- Filter ((timestamp#1-T3000ms >= window#859-T3000ms.start) AND (timestamp#1-T3000ms < window#859-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#136, hostname#140, time#145, method#151, url#158, responsecode#166, bytes#175), 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#136, hostname#140, time#145, method#151, url#158, responsecode#166, bytes#175)], [window#859-T3000ms, logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, url#158, responsecode#166, bytes#175]
            +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, url#158, responsecode#166, split(logs#5, ,, -1)[6] AS bytes#175]
               +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, url#158, split(logs#5, ,, -1)[5] AS responsecode#166]
                  +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, split(logs#5, ,, -1)[4] AS url#158]
                     +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, split(logs#5, ,, -1)[3] AS method#151]
                        +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, split(logs#5, ,, -1)[2] AS time#145]
                           +- Project [logs#5, timestamp#1-T3000ms, idx#136, split(logs#5, ,, -1)[1] AS hostname#140]
                              +- Project [logs#5, timestamp#1-T3000ms, split(logs#5, ,, -1)[0] AS idx#136]
                                 +- 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@6b3ca49a, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 16, 32

== Analyzed Logical Plan ==

WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@702a5285
+- Sort [total_bytes#858 DESC NULLS LAST], true
   +- Aggregate [window#859-T3000ms, hostname#140], [window#859-T3000ms AS window#847-T3000ms, hostname#140, sum(cast(bytes#175 as double)) AS total_bytes#858]
      +- Filter ((timestamp#1-T3000ms >= window#859-T3000ms.start) AND (timestamp#1-T3000ms < window#859-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#136, hostname#140, time#145, method#151, url#158, responsecode#166, bytes#175), 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#136, hostname#140, time#145, method#151, url#158, responsecode#166, bytes#175)], [window#859-T3000ms, logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, url#158, responsecode#166, bytes#175]
            +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, url#158, responsecode#166, split(logs#5, ,, -1)[6] AS bytes#175]
               +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, url#158, split(logs#5, ,, -1)[5] AS responsecode#166]
                  +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, method#151, split(logs#5, ,, -1)[4] AS url#158]
                     +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, time#145, split(logs#5, ,, -1)[3] AS method#151]
                        +- Project [logs#5, timestamp#1-T3000ms, idx#136, hostname#140, split(logs#5, ,, -1)[2] AS time#145]
                           +- Project [logs#5, timestamp#1-T3000ms, idx#136, split(logs#5, ,, -1)[1] AS hostname#140]
                              +- Project [logs#5, timestamp#1-T3000ms, split(logs#5, ,, -1)[0] AS idx#136]
                                 +- 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@6b3ca49a, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 16, 32

== Optimized Logical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@702a5285
+- Sort [total_bytes#858 DESC NULLS LAST], true
   +- Aggregate [window#859-T3000ms, hostname#140], [window#859-T3000ms AS window#847-T3000ms, hostname#140, sum(cast(bytes#175 as double)) AS total_bytes#858]
      +- Project [window#859-T3000ms, hostname#140, bytes#175]
         +- Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#859-T3000ms)) AND (timestamp#1-T3000ms >= window#859-T3000ms.start)) AND (timestamp#1-T3000ms < window#859-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#140, bytes#175), 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#140, bytes#175)], [window#859-T3000ms, timestamp#1-T3000ms, hostname#140, bytes#175]
               +- Project [timestamp#1-T3000ms, split(logs#5, ,, -1)[1] AS hostname#140, split(logs#5, ,, -1)[6] AS bytes#175]
                  +- 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@6b3ca49a, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 16, 32

== Physical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@702a5285
+- *(7) Sort [total_bytes#858 DESC NULLS LAST], true, 0
   +- Exchange rangepartitioning(total_bytes#858 DESC NULLS LAST, 200), true, [id=#1212]
      +- *(6) HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[sum(cast(bytes#175 as double))], output=[window#847-T3000ms, hostname#140, total_bytes#858])
         +- StateStoreSave [window#859-T3000ms, hostname#140], state info [ checkpoint = file:/tmp/temporary-fcb72abb-9143-4d38-a65d-25baf17e34c7/state, runId = 49431645-20e3-4737-9dcc-8f334da22f1d, opId = 0, ver = 2, numPartitions = 200], Complete, 1733343553054, 2
            +- *(5) HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[merge_sum(cast(bytes#175 as double))], output=[window#859-T3000ms, hostname#140, sum#869])
               +- StateStoreRestore [window#859-T3000ms, hostname#140], state info [ checkpoint = file:/tmp/temporary-fcb72abb-9143-4d38-a65d-25baf17e34c7/state, runId = 49431645-20e3-4737-9dcc-8f334da22f1d, opId = 0, ver = 2, numPartitions = 200], 2
                  +- *(4) HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[merge_sum(cast(bytes#175 as double))], output=[window#859-T3000ms, hostname#140, sum#869])
                     +- Exchange hashpartitioning(window#859-T3000ms, hostname#140, 200), true, [id=#1200]
                        +- *(3) HashAggregate(keys=[window#859-T3000ms, hostname#140], functions=[partial_sum(cast(bytes#175 as double))], output=[window#859-T3000ms, hostname#140, sum#869])
                           +- *(3) Project [window#859-T3000ms, hostname#140, bytes#175]
                              +- *(3) Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#859-T3000ms)) AND (timestamp#1-T3000ms >= window#859-T3000ms.start)) AND (timestamp#1-T3000ms < window#859-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#140, bytes#175), 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#140, bytes#175)], [window#859-T3000ms, timestamp#1-T3000ms, hostname#140, bytes#175]
                                    +- *(3) Project [timestamp#1-T3000ms, split(logs#5, ,, -1)[1] AS hostname#140, split(logs#5, ,, -1)[6] AS bytes#175]
                                       +- 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