NasaLogSparkStreaming - Details for Query 27

Details for Query 27

Submitted Time: 2024/12/27 14:33:16
Duration: 21 s
Succeeded Jobs: 12

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 (6)\n \nduration: total (min, med, max (stageId: taskId))\n9.5 s (30 ms, 36 ms, 116 ms (stage 21.0: task 1222))"; 2 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 9.4 s (29 ms, 36 ms, 114 ms (stage 21.0: task 1222)) peak memory total (min, med, max (stageId: taskId)) 242.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 21.0: task 1222)) number of output rows: 3 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 21.0: task 1222))"]; } 3 [labelType="html" label="StateStoreSave number of total state rows: 3 memory used by state total (min, med, max (stageId: taskId)) 82.0 KiB (416.0 B, 416.0 B, 720.0 B (stage 21.0: task 1222)) count of cache hit on states cache in provider: 800 number of output rows: 3 estimated size of state only on current version total (min, med, max (stageId: taskId)) 20.9 KiB (104.0 B, 104.0 B, 304.0 B (stage 21.0: task 1222)) time to commit changes total (min, med, max (stageId: taskId)) 9.0 s (26 ms, 35 ms, 96 ms (stage 21.0: task 1257)) time to remove total (min, med, max (stageId: taskId)) 0 ms (0 ms, 0 ms, 0 ms (stage 21.0: task 1205)) number of updated state rows: 3 time to update total (min, med, max (stageId: taskId)) 412 ms (0 ms, 1 ms, 62 ms (stage 21.0: task 1222))"]; subgraph cluster4 { isCluster="true"; label="WholeStageCodegen (5)\n \nduration: total (min, med, max (stageId: taskId))\n487 ms (0 ms, 1 ms, 63 ms (stage 21.0: task 1222))"; 5 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 106 ms (0 ms, 0 ms, 42 ms (stage 21.0: task 1222)) peak memory total (min, med, max (stageId: taskId)) 242.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 21.0: task 1222)) number of output rows: 3 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 21.0: task 1222))"]; } 6 [labelType="html" label="StateStoreRestore number of output rows: 3"]; subgraph cluster7 { isCluster="true"; label="WholeStageCodegen (4)\n \nduration: total (min, med, max (stageId: taskId))\n782 ms (1 ms, 2 ms, 56 ms (stage 21.0: task 1370))"; 8 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 16 ms (0 ms, 0 ms, 15 ms (stage 21.0: task 1370)) peak memory total (min, med, max (stageId: taskId)) 242.0 MiB (256.0 KiB, 256.0 KiB, 64.3 MiB (stage 21.0: task 1222)) number of output rows: 3 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 21.0: task 1222))"]; } 9 [labelType="html" label="Exchange shuffle records written: 6 shuffle write time total (min, med, max (stageId: taskId)) 23 ms (5 ms, 17 ms, 17 ms (stage 20.0: task 1197)) records read: 6 local bytes read total (min, med, max (stageId: taskId)) 314.0 B (0.0 B, 0.0 B, 158.0 B (stage 21.0: task 1222)) fetch wait time total (min, med, max (stageId: taskId)) 13 ms (0 ms, 0 ms, 13 ms (stage 21.0: task 1370)) remote bytes read total (min, med, max (stageId: taskId)) 156.0 B (0.0 B, 0.0 B, 156.0 B (stage 21.0: task 1370)) local blocks read: 4 remote blocks read: 2 data size total (min, med, max (stageId: taskId)) 288.0 B (144.0 B, 144.0 B, 144.0 B (stage 20.0: task 1197)) shuffle bytes written total (min, med, max (stageId: taskId)) 470.0 B (235.0 B, 235.0 B, 235.0 B (stage 20.0: task 1197))"]; subgraph cluster10 { isCluster="true"; label="WholeStageCodegen (3)\n \nduration: total (min, med, max (stageId: taskId))\n36 ms (8 ms, 28 ms, 28 ms (stage 20.0: task 1197))"; 11 [labelType="html" label="HashAggregate time in aggregation build total (min, med, max (stageId: taskId)) 3 ms (1 ms, 2 ms, 2 ms (stage 20.0: task 1197)) peak memory total (min, med, max (stageId: taskId)) 128.5 MiB (64.3 MiB, 64.3 MiB, 64.3 MiB (stage 20.0: task 1197)) number of output rows: 6 avg hash probe bucket list iters (min, med, max (stageId: taskId)): (1, 1, 1 (stage 20.0: task 1197))"]; 12 [labelType="html" label=" Project "]; 13 [labelType="html" label="Filter number of output rows: 30"]; 14 [labelType="html" label="Expand number of output rows: 30"]; } 15 [labelType="html" label=" EventTimeWatermark "]; subgraph cluster16 { isCluster="true"; label="WholeStageCodegen (2)\n \nduration: total (min, med, max (stageId: taskId))\n52 ms (4 ms, 48 ms, 48 ms (stage 20.0: task 1197))"; 17 [labelType="html" label=" Project "]; 18 [labelType="html" label="Filter number of output rows: 15"]; } 19 [labelType="html" label="Generate number of output rows: 23"]; subgraph cluster20 { isCluster="true"; label="WholeStageCodegen (1)\n \nduration: total (min, med, max (stageId: taskId))\n70 ms (5 ms, 65 ms, 65 ms (stage 20.0: task 1197))"; 21 [labelType="html" label=" Project "]; } 22 [labelType="html" label="MicroBatchScan number of output rows: 23"]; 2->0; 3->2; 5->3; 6->5; 8->6; 9->8; 11->9; 12->11; 13->12; 14->13; 15->14; 17->15; 18->17; 19->18; 21->19; 22->21; }

WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@a631a1f

HashAggregate(keys=[window#81-T3000ms], functions=[count(1)])

WholeStageCodegen (6)

StateStoreSave [window#81-T3000ms], state info [ checkpoint = file:/tmp/temporary-44a37146-c22b-4332-803b-47058dc7d369/state, runId = 2ad6ec75-02a9-40e4-8f05-f0dd3dbf4b8d, opId = 0, ver = 2, numPartitions = 200], Update, 1735309969784, 2

HashAggregate(keys=[window#81-T3000ms], functions=[merge_count(1)])

WholeStageCodegen (5)

StateStoreRestore [window#81-T3000ms], state info [ checkpoint = file:/tmp/temporary-44a37146-c22b-4332-803b-47058dc7d369/state, runId = 2ad6ec75-02a9-40e4-8f05-f0dd3dbf4b8d, opId = 0, ver = 2, numPartitions = 200], 2

HashAggregate(keys=[window#81-T3000ms], functions=[merge_count(1)])

WholeStageCodegen (4)

Exchange hashpartitioning(window#81-T3000ms, 200), true, [id=#685]

HashAggregate(keys=[window#81-T3000ms], functions=[partial_count(1)])

Project [window#81-T3000ms]

Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#81-T3000ms)) AND (timestamp#1-T3000ms >= window#81-T3000ms.start)) AND (timestamp#1-T3000ms < window#81-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), 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)], [window#81-T3000ms, timestamp#1-T3000ms]

WholeStageCodegen (3)

EventTimeWatermark timestamp#1: timestamp, 3 seconds

Project [timestamp#1]

Filter Contains(split(logs#5, ,, -1)[4], gif)

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@a631a1f
+- SubqueryAlias gif_count
   +- Aggregate [window#81-T3000ms], [window#81-T3000ms AS window#69-T3000ms, count(1) AS count#80L]
      +- Filter ((timestamp#1-T3000ms >= window#81-T3000ms.start) AND (timestamp#1-T3000ms < window#81-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#10, hostname#14, time#19, method#25, url#32, responsecode#40, bytes#49), 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#10, hostname#14, time#19, method#25, url#32, responsecode#40, bytes#49)], [window#81-T3000ms, logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, url#32, responsecode#40, bytes#49]
            +- Filter Contains(url#32, gif)
               +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, url#32, responsecode#40, split(logs#5, ,, -1)[6] AS bytes#49]
                  +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, url#32, split(logs#5, ,, -1)[5] AS responsecode#40]
                     +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, split(logs#5, ,, -1)[4] AS url#32]
                        +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, split(logs#5, ,, -1)[3] AS method#25]
                           +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, split(logs#5, ,, -1)[2] AS time#19]
                              +- Project [logs#5, timestamp#1-T3000ms, idx#10, split(logs#5, ,, -1)[1] AS hostname#14]
                                 +- Project [logs#5, timestamp#1-T3000ms, split(logs#5, ,, -1)[0] AS idx#10]
                                    +- 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@466d1a44, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 10, 33

== Analyzed Logical Plan ==

WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@a631a1f
+- SubqueryAlias gif_count
   +- Aggregate [window#81-T3000ms], [window#81-T3000ms AS window#69-T3000ms, count(1) AS count#80L]
      +- Filter ((timestamp#1-T3000ms >= window#81-T3000ms.start) AND (timestamp#1-T3000ms < window#81-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#10, hostname#14, time#19, method#25, url#32, responsecode#40, bytes#49), 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#10, hostname#14, time#19, method#25, url#32, responsecode#40, bytes#49)], [window#81-T3000ms, logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, url#32, responsecode#40, bytes#49]
            +- Filter Contains(url#32, gif)
               +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, url#32, responsecode#40, split(logs#5, ,, -1)[6] AS bytes#49]
                  +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, url#32, split(logs#5, ,, -1)[5] AS responsecode#40]
                     +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, method#25, split(logs#5, ,, -1)[4] AS url#32]
                        +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, time#19, split(logs#5, ,, -1)[3] AS method#25]
                           +- Project [logs#5, timestamp#1-T3000ms, idx#10, hostname#14, split(logs#5, ,, -1)[2] AS time#19]
                              +- Project [logs#5, timestamp#1-T3000ms, idx#10, split(logs#5, ,, -1)[1] AS hostname#14]
                                 +- Project [logs#5, timestamp#1-T3000ms, split(logs#5, ,, -1)[0] AS idx#10]
                                    +- 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@466d1a44, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 10, 33

== Optimized Logical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@a631a1f
+- Aggregate [window#81-T3000ms], [window#81-T3000ms AS window#69-T3000ms, count(1) AS count#80L]
   +- Project [window#81-T3000ms]
      +- Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#81-T3000ms)) AND (timestamp#1-T3000ms >= window#81-T3000ms.start)) AND (timestamp#1-T3000ms < window#81-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), 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)], [window#81-T3000ms, timestamp#1-T3000ms]
            +- EventTimeWatermark timestamp#1: timestamp, 3 seconds
               +- Project [timestamp#1]
                  +- Filter Contains(split(logs#5, ,, -1)[4], gif)
                     +- 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@466d1a44, TextSocketV2[host: stream-emulator.data-science-tools.svc.cluster.local, port: 5551], 10, 33

== Physical Plan ==
WriteToDataSourceV2 org.apache.spark.sql.execution.streaming.sources.MicroBatchWrite@a631a1f
+- *(6) HashAggregate(keys=[window#81-T3000ms], functions=[count(1)], output=[window#69-T3000ms, count#80L])
   +- StateStoreSave [window#81-T3000ms], state info [ checkpoint = file:/tmp/temporary-44a37146-c22b-4332-803b-47058dc7d369/state, runId = 2ad6ec75-02a9-40e4-8f05-f0dd3dbf4b8d, opId = 0, ver = 2, numPartitions = 200], Update, 1735309969784, 2
      +- *(5) HashAggregate(keys=[window#81-T3000ms], functions=[merge_count(1)], output=[window#81-T3000ms, count#102L])
         +- StateStoreRestore [window#81-T3000ms], state info [ checkpoint = file:/tmp/temporary-44a37146-c22b-4332-803b-47058dc7d369/state, runId = 2ad6ec75-02a9-40e4-8f05-f0dd3dbf4b8d, opId = 0, ver = 2, numPartitions = 200], 2
            +- *(4) HashAggregate(keys=[window#81-T3000ms], functions=[merge_count(1)], output=[window#81-T3000ms, count#102L])
               +- Exchange hashpartitioning(window#81-T3000ms, 200), true, [id=#685]
                  +- *(3) HashAggregate(keys=[window#81-T3000ms], functions=[partial_count(1)], output=[window#81-T3000ms, count#102L])
                     +- *(3) Project [window#81-T3000ms]
                        +- *(3) Filter (((isnotnull(timestamp#1-T3000ms) AND isnotnull(window#81-T3000ms)) AND (timestamp#1-T3000ms >= window#81-T3000ms.start)) AND (timestamp#1-T3000ms < window#81-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), 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)], [window#81-T3000ms, timestamp#1-T3000ms]
                              +- EventTimeWatermark timestamp#1: timestamp, 3 seconds
                                 +- *(2) Project [timestamp#1]
                                    +- *(2) Filter Contains(split(logs#5, ,, -1)[4], gif)
                                       +- 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