http://git-wip-us.apache.org/repos/asf/hive/blob/16d28b34/ql/src/test/results/clientpositive/spark/vector_data_types.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/vector_data_types.q.out b/ql/src/test/results/clientpositive/spark/vector_data_types.q.out index dbaf14d..79638c1 100644 --- a/ql/src/test/results/clientpositive/spark/vector_data_types.q.out +++ b/ql/src/test/results/clientpositive/spark/vector_data_types.q.out @@ -97,10 +97,14 @@ POSTHOOK: Lineage: over1korc.s SIMPLE [(over1k)over1k.FieldSchema(name:s, type:s POSTHOOK: Lineage: over1korc.si SIMPLE [(over1k)over1k.FieldSchema(name:si, type:smallint, comment:null), ] POSTHOOK: Lineage: over1korc.t SIMPLE [(over1k)over1k.FieldSchema(name:t, type:tinyint, comment:null), ] POSTHOOK: Lineage: over1korc.ts SIMPLE [(over1k)over1k.FieldSchema(name:ts, type:timestamp, comment:null), ] -PREHOOK: query: EXPLAIN SELECT t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 +PREHOOK: query: EXPLAIN VECTORIZATION EXPRESSION SELECT t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 PREHOOK: type: QUERY -POSTHOOK: query: EXPLAIN SELECT t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 +POSTHOOK: query: EXPLAIN VECTORIZATION EXPRESSION SELECT t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: false + enabledConditionsNotMet: [hive.vectorized.execution.enabled IS false] + STAGE DEPENDENCIES: Stage-1 is a root stage Stage-0 depends on stages: Stage-1 @@ -189,10 +193,14 @@ POSTHOOK: type: QUERY POSTHOOK: Input: default@over1korc #### A masked pattern was here #### -17045922556 -PREHOOK: query: EXPLAIN select t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 +PREHOOK: query: EXPLAIN VECTORIZATION EXPRESSION select t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 PREHOOK: type: QUERY -POSTHOOK: query: EXPLAIN select t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 +POSTHOOK: query: EXPLAIN VECTORIZATION EXPRESSION select t, si, i, b, f, d, bo, s, ts, dec, bin FROM over1korc ORDER BY t, si, i LIMIT 20 POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: Stage-1 is a root stage Stage-0 depends on stages: Stage-1 @@ -209,29 +217,66 @@ STAGE PLANS: TableScan alias: over1korc Statistics: Num rows: 1049 Data size: 311170 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] Select Operator expressions: t (type: tinyint), si (type: smallint), i (type: int), b (type: bigint), f (type: float), d (type: double), bo (type: boolean), s (type: string), ts (type: timestamp), dec (type: decimal(4,2)), bin (type: binary) outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] Statistics: Num rows: 1049 Data size: 311170 Basic stats: COMPLETE Column stats: NONE Reduce Output Operator key expressions: _col0 (type: tinyint), _col1 (type: smallint), _col2 (type: int) sort order: +++ + Reduce Sink Vectorization: + className: VectorReduceSinkOperator + native: false + nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true, Not ACID UPDATE or DELETE IS true, No buckets IS true, No DISTINCT columns IS true, BinarySortableSerDe for keys IS true, LazyBinarySerDe for values IS true + nativeConditionsNotMet: No TopN IS false, Uniform Hash IS false Statistics: Num rows: 1049 Data size: 311170 Basic stats: COMPLETE Column stats: NONE TopN Hash Memory Usage: 0.1 value expressions: _col3 (type: bigint), _col4 (type: float), _col5 (type: double), _col6 (type: boolean), _col7 (type: string), _col8 (type: timestamp), _col9 (type: decimal(4,2)), _col10 (type: binary) Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reducer 2 Execution mode: vectorized + Reduce Vectorization: + enabled: true + enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true + groupByVectorOutput: true + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reduce Operator Tree: Select Operator expressions: KEY.reducesinkkey0 (type: tinyint), KEY.reducesinkkey1 (type: smallint), KEY.reducesinkkey2 (type: int), VALUE._col0 (type: bigint), VALUE._col1 (type: float), VALUE._col2 (type: double), VALUE._col3 (type: boolean), VALUE._col4 (type: string), VALUE._col5 (type: timestamp), VALUE._col6 (type: decimal(4,2)), VALUE._col7 (type: binary) outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10] Statistics: Num rows: 1049 Data size: 311170 Basic stats: COMPLETE Column stats: NONE Limit Number of rows: 20 + Limit Vectorization: + className: VectorLimitOperator + native: true Statistics: Num rows: 20 Data size: 5920 Basic stats: COMPLETE Column stats: NONE File Output Operator compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false Statistics: Num rows: 20 Data size: 5920 Basic stats: COMPLETE Column stats: NONE table: input format: org.apache.hadoop.mapred.SequenceFileInputFormat
http://git-wip-us.apache.org/repos/asf/hive/blob/16d28b34/ql/src/test/results/clientpositive/spark/vector_decimal_aggregate.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/vector_decimal_aggregate.q.out b/ql/src/test/results/clientpositive/spark/vector_decimal_aggregate.q.out index cfdfce1..919e290 100644 --- a/ql/src/test/results/clientpositive/spark/vector_decimal_aggregate.q.out +++ b/ql/src/test/results/clientpositive/spark/vector_decimal_aggregate.q.out @@ -23,7 +23,7 @@ POSTHOOK: Lineage: decimal_vgby.cint SIMPLE [(alltypesorc)alltypesorc.FieldSchem PREHOOK: query: -- SORT_QUERY_RESULTS -- First only do simple aggregations that output primitives only -EXPLAIN SELECT cint, +EXPLAIN VECTORIZATION EXPRESSION SELECT cint, COUNT(cdecimal1), MAX(cdecimal1), MIN(cdecimal1), SUM(cdecimal1), COUNT(cdecimal2), MAX(cdecimal2), MIN(cdecimal2), SUM(cdecimal2) FROM decimal_vgby @@ -33,13 +33,17 @@ PREHOOK: type: QUERY POSTHOOK: query: -- SORT_QUERY_RESULTS -- First only do simple aggregations that output primitives only -EXPLAIN SELECT cint, +EXPLAIN VECTORIZATION EXPRESSION SELECT cint, COUNT(cdecimal1), MAX(cdecimal1), MIN(cdecimal1), SUM(cdecimal1), COUNT(cdecimal2), MAX(cdecimal2), MIN(cdecimal2), SUM(cdecimal2) FROM decimal_vgby GROUP BY cint HAVING COUNT(*) > 1 POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: Stage-1 is a root stage Stage-0 depends on stages: Stage-1 @@ -56,12 +60,26 @@ STAGE PLANS: TableScan alias: decimal_vgby Statistics: Num rows: 12288 Data size: 2165060 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3] Select Operator expressions: cint (type: int), cdecimal1 (type: decimal(20,10)), cdecimal2 (type: decimal(23,14)) outputColumnNames: cint, cdecimal1, cdecimal2 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [3, 1, 2] Statistics: Num rows: 12288 Data size: 2165060 Basic stats: COMPLETE Column stats: NONE Group By Operator aggregations: count(cdecimal1), max(cdecimal1), min(cdecimal1), sum(cdecimal1), count(cdecimal2), max(cdecimal2), min(cdecimal2), sum(cdecimal2), count() + Group By Vectorization: + aggregators: VectorUDAFCount(col 1) -> bigint, VectorUDAFMaxDecimal(col 1) -> decimal(20,10), VectorUDAFMinDecimal(col 1) -> decimal(20,10), VectorUDAFSumDecimal(col 1) -> decimal(38,18), VectorUDAFCount(col 2) -> bigint, VectorUDAFMaxDecimal(col 2) -> decimal(23,14), VectorUDAFMinDecimal(col 2) -> decimal(23,14), VectorUDAFSumDecimal(col 2) -> decimal(38,18), VectorUDAFCountStar(*) -> bigint + className: VectorGroupByOperator + vectorOutput: true + keyExpressions: col 3 + native: false + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8] keys: cint (type: int) mode: hash outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9 @@ -70,27 +88,65 @@ STAGE PLANS: key expressions: _col0 (type: int) sort order: + Map-reduce partition columns: _col0 (type: int) + Reduce Sink Vectorization: + className: VectorReduceSinkOperator + native: false + nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true, Not ACID UPDATE or DELETE IS true, No buckets IS true, No TopN IS true, No DISTINCT columns IS true, BinarySortableSerDe for keys IS true, LazyBinarySerDe for values IS true + nativeConditionsNotMet: Uniform Hash IS false Statistics: Num rows: 12288 Data size: 2165060 Basic stats: COMPLETE Column stats: NONE value expressions: _col1 (type: bigint), _col2 (type: decimal(20,10)), _col3 (type: decimal(20,10)), _col4 (type: decimal(30,10)), _col5 (type: bigint), _col6 (type: decimal(23,14)), _col7 (type: decimal(23,14)), _col8 (type: decimal(33,14)), _col9 (type: bigint) Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reducer 2 Execution mode: vectorized + Reduce Vectorization: + enabled: true + enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true + groupByVectorOutput: true + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reduce Operator Tree: Group By Operator aggregations: count(VALUE._col0), max(VALUE._col1), min(VALUE._col2), sum(VALUE._col3), count(VALUE._col4), max(VALUE._col5), min(VALUE._col6), sum(VALUE._col7), count(VALUE._col8) + Group By Vectorization: + aggregators: VectorUDAFCountMerge(col 1) -> bigint, VectorUDAFMaxDecimal(col 2) -> decimal(20,10), VectorUDAFMinDecimal(col 3) -> decimal(20,10), VectorUDAFSumDecimal(col 4) -> decimal(38,18), VectorUDAFCountMerge(col 5) -> bigint, VectorUDAFMaxDecimal(col 6) -> decimal(23,14), VectorUDAFMinDecimal(col 7) -> decimal(23,14), VectorUDAFSumDecimal(col 8) -> decimal(38,18), VectorUDAFCountMerge(col 9) -> bigint + className: VectorGroupByOperator + vectorOutput: true + keyExpressions: col 0 + native: false + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8] keys: KEY._col0 (type: int) mode: mergepartial outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9 Statistics: Num rows: 6144 Data size: 1082530 Basic stats: COMPLETE Column stats: NONE Filter Operator + Filter Vectorization: + className: VectorFilterOperator + native: true + predicateExpression: FilterLongColGreaterLongScalar(col 9, val 1) -> boolean predicate: (_col9 > 1) (type: boolean) Statistics: Num rows: 2048 Data size: 360843 Basic stats: COMPLETE Column stats: NONE Select Operator expressions: _col0 (type: int), _col1 (type: bigint), _col2 (type: decimal(20,10)), _col3 (type: decimal(20,10)), _col4 (type: decimal(30,10)), _col5 (type: bigint), _col6 (type: decimal(23,14)), _col7 (type: decimal(23,14)), _col8 (type: decimal(33,14)) outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8] Statistics: Num rows: 2048 Data size: 360843 Basic stats: COMPLETE Column stats: NONE File Output Operator compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false Statistics: Num rows: 2048 Data size: 360843 Basic stats: COMPLETE Column stats: NONE table: input format: org.apache.hadoop.mapred.SequenceFileInputFormat @@ -130,7 +186,7 @@ POSTHOOK: Input: default@decimal_vgby 762 2 5831542.2692483780 1531.2194054054 5833073.4886537834 2 6984454.21109769200000 1833.94569230769250 6986288.15678999969250 NULL 3072 9318.4351351351 -4298.1513513514 5018444.1081079808 3072 11160.71538461538500 -5147.90769230769300 6010604.30769230735360 PREHOOK: query: -- Now add the others... -EXPLAIN SELECT cint, +EXPLAIN VECTORIZATION EXPRESSION SELECT cint, COUNT(cdecimal1), MAX(cdecimal1), MIN(cdecimal1), SUM(cdecimal1), AVG(cdecimal1), STDDEV_POP(cdecimal1), STDDEV_SAMP(cdecimal1), COUNT(cdecimal2), MAX(cdecimal2), MIN(cdecimal2), SUM(cdecimal2), AVG(cdecimal2), STDDEV_POP(cdecimal2), STDDEV_SAMP(cdecimal2) FROM decimal_vgby @@ -138,13 +194,17 @@ EXPLAIN SELECT cint, HAVING COUNT(*) > 1 PREHOOK: type: QUERY POSTHOOK: query: -- Now add the others... -EXPLAIN SELECT cint, +EXPLAIN VECTORIZATION EXPRESSION SELECT cint, COUNT(cdecimal1), MAX(cdecimal1), MIN(cdecimal1), SUM(cdecimal1), AVG(cdecimal1), STDDEV_POP(cdecimal1), STDDEV_SAMP(cdecimal1), COUNT(cdecimal2), MAX(cdecimal2), MIN(cdecimal2), SUM(cdecimal2), AVG(cdecimal2), STDDEV_POP(cdecimal2), STDDEV_SAMP(cdecimal2) FROM decimal_vgby GROUP BY cint HAVING COUNT(*) > 1 POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: Stage-1 is a root stage Stage-0 depends on stages: Stage-1 @@ -161,12 +221,27 @@ STAGE PLANS: TableScan alias: decimal_vgby Statistics: Num rows: 12288 Data size: 2165060 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3] Select Operator expressions: cint (type: int), cdecimal1 (type: decimal(20,10)), cdecimal2 (type: decimal(23,14)) outputColumnNames: cint, cdecimal1, cdecimal2 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [3, 1, 2] Statistics: Num rows: 12288 Data size: 2165060 Basic stats: COMPLETE Column stats: NONE Group By Operator aggregations: count(cdecimal1), max(cdecimal1), min(cdecimal1), sum(cdecimal1), avg(cdecimal1), stddev_pop(cdecimal1), stddev_samp(cdecimal1), count(cdecimal2), max(cdecimal2), min(cdecimal2), sum(cdecimal2), avg(cdecimal2), stddev_pop(cdecimal2), stddev_samp(cdecimal2), count() + Group By Vectorization: + aggregators: VectorUDAFCount(col 1) -> bigint, VectorUDAFMaxDecimal(col 1) -> decimal(20,10), VectorUDAFMinDecimal(col 1) -> decimal(20,10), VectorUDAFSumDecimal(col 1) -> decimal(38,18), VectorUDAFAvgDecimal(col 1) -> struct<count:bigint,sum:decimal(30,10)>, VectorUDAFStdPopDecimal(col 1) -> struct<count:bigint,sum:double,variance:double>, VectorUDAFStdSampDecimal(col 1) -> struct<count:bigint,sum:double,variance:double>, VectorUDAFCount(col 2) -> bigint, VectorUDAFMaxDecimal(col 2) -> decimal(23,14), VectorUDAFMinDecimal(col 2) -> decimal(23,14), VectorUDAFSumDecimal(col 2) -> decimal(38,18), VectorUDAFAvgDecimal(col 2) -> struct<count:bigint,sum:decimal(33,14)>, VectorUDAFStdPopDecimal(col 2) -> struct<count:bigint,sum:double,variance:double>, VectorUDAFStdSampDecimal(col 2) -> struct<count:bigint,sum:double,variance:double>, VectorUDAFCountStar(*) -> bigint + className: VectorGroupByOperator + vectorOutput: false + keyExpressions: col 3 + native: false + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14] + vectorOutputConditionsNotMet: Vector output of VectorUDAFAvgDecimal(col 1) -> struct<count:bigint,sum:decimal(30,10)> output type STRUCT requires PRIMITIVE IS false, Vector output of VectorUDAFStdPopDecimal(col 1) -> struct<count:bigint,sum:double,variance:double> output type STRUCT requires PRIMITIVE IS false, Vector output of VectorUDAFStdSampDecimal(col 1) -> struct<count:bigint,sum:double,variance:double> output type STRUCT requires PRIMITIVE IS false, Vector output of VectorUDAFAvgDecimal(col 2) -> struct<count:bigint,sum:decimal(33,14)> output type STRUCT requires PRIMITIVE IS false, Vector output of VectorUDAFStdPopDecimal(col 2) -> struct<count:bigint,sum:double,variance:double> output type STRUCT requires PRIMITIVE IS false, Vector output of VectorUDAFStdSampDecimal(col 2) -> struct<count:bigint,sum:double,variance:double> output type STRUCT requires PRIMITIVE IS false keys: cint (type: int) mode: hash outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9, _col10, _col11, _col12, _col13, _col14, _col15 @@ -178,7 +253,20 @@ STAGE PLANS: Statistics: Num rows: 12288 Data size: 2165060 Basic stats: COMPLETE Column stats: NONE value expressions: _col1 (type: bigint), _col2 (type: decimal(20,10)), _col3 (type: decimal(20,10)), _col4 (type: decimal(30,10)), _col5 (type: struct<count:bigint,sum:decimal(30,10),input:decimal(20,10)>), _col6 (type: struct<count:bigint,sum:double,variance:double>), _col7 (type: struct<count:bigint,sum:double,variance:double>), _col8 (type: bigint), _col9 (type: decimal(23,14)), _col10 (type: decimal(23,14)), _col11 (type: decimal(33,14)), _col12 (type: struct<count:bigint,sum:decimal(33,14),input:decimal(23,14)>), _col13 (type: struct<count:bigint,sum:double,variance:double>), _col14 (type: struct<count:bigint,sum:double,variance:double>), _col15 (type: bigint) Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: false + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reducer 2 + Reduce Vectorization: + enabled: true + enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true + notVectorizedReason: Aggregation Function UDF avg parameter expression for GROUPBY operator: Data type struct<count:bigint,sum:decimal(30,10),input:decimal(20,10)> of Column[VALUE._col4] not supported + vectorized: false Reduce Operator Tree: Group By Operator aggregations: count(VALUE._col0), max(VALUE._col1), min(VALUE._col2), sum(VALUE._col3), avg(VALUE._col4), stddev_pop(VALUE._col5), stddev_samp(VALUE._col6), count(VALUE._col7), max(VALUE._col8), min(VALUE._col9), sum(VALUE._col10), avg(VALUE._col11), stddev_pop(VALUE._col12), stddev_samp(VALUE._col13), count(VALUE._col14) http://git-wip-us.apache.org/repos/asf/hive/blob/16d28b34/ql/src/test/results/clientpositive/spark/vector_decimal_mapjoin.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/vector_decimal_mapjoin.q.out b/ql/src/test/results/clientpositive/spark/vector_decimal_mapjoin.q.out index 0493994..11d7609 100644 --- a/ql/src/test/results/clientpositive/spark/vector_decimal_mapjoin.q.out +++ b/ql/src/test/results/clientpositive/spark/vector_decimal_mapjoin.q.out @@ -72,12 +72,16 @@ POSTHOOK: type: QUERY POSTHOOK: Input: default@over1k POSTHOOK: Output: default@t2 POSTHOOK: Lineage: t2.dec EXPRESSION [(over1k)over1k.FieldSchema(name:dec, type:decimal(4,2), comment:null), ] -PREHOOK: query: explain +PREHOOK: query: explain vectorization expression select t1.dec, t2.dec from t1 join t2 on (t1.dec=t2.dec) PREHOOK: type: QUERY -POSTHOOK: query: explain +POSTHOOK: query: explain vectorization expression select t1.dec, t2.dec from t1 join t2 on (t1.dec=t2.dec) POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: Stage-2 is a root stage Stage-1 depends on stages: Stage-2 @@ -93,18 +97,40 @@ STAGE PLANS: TableScan alias: t2 Statistics: Num rows: 1049 Data size: 117488 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0] Filter Operator + Filter Vectorization: + className: VectorFilterOperator + native: true + predicateExpression: SelectColumnIsNotNull(col 1)(children: CastDecimalToBoolean(col 0) -> 1:long) -> boolean predicate: dec is not null (type: boolean) Statistics: Num rows: 1049 Data size: 117488 Basic stats: COMPLETE Column stats: NONE Select Operator expressions: dec (type: decimal(4,0)) outputColumnNames: _col0 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0] Statistics: Num rows: 1049 Data size: 117488 Basic stats: COMPLETE Column stats: NONE Spark HashTable Sink Operator + Spark Hash Table Sink Vectorization: + className: VectorSparkHashTableSinkOperator + native: true keys: 0 _col0 (type: decimal(6,2)) 1 _col0 (type: decimal(6,2)) Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: true + usesVectorUDFAdaptor: false + vectorized: true Local Work: Map Reduce Local Work @@ -117,12 +143,23 @@ STAGE PLANS: TableScan alias: t1 Statistics: Num rows: 1049 Data size: 117488 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0] Filter Operator + Filter Vectorization: + className: VectorFilterOperator + native: true + predicateExpression: SelectColumnIsNotNull(col 1)(children: CastDecimalToBoolean(col 0) -> 1:long) -> boolean predicate: dec is not null (type: boolean) Statistics: Num rows: 1049 Data size: 117488 Basic stats: COMPLETE Column stats: NONE Select Operator expressions: dec (type: decimal(4,2)) outputColumnNames: _col0 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0] Statistics: Num rows: 1049 Data size: 117488 Basic stats: COMPLETE Column stats: NONE Map Join Operator condition map: @@ -130,18 +167,35 @@ STAGE PLANS: keys: 0 _col0 (type: decimal(6,2)) 1 _col0 (type: decimal(6,2)) + Map Join Vectorization: + className: VectorMapJoinOperator + native: false + nativeConditionsMet: hive.vectorized.execution.mapjoin.native.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true, One MapJoin Condition IS true, No nullsafe IS true, Not empty key IS true, When Fast Hash Table, then requires no Hybrid Hash Join IS true, Small table vectorizes IS true + nativeConditionsNotMet: Supports Key Types IS false + nativeNotSupportedKeyTypes: DECIMAL outputColumnNames: _col0, _col1 input vertices: 1 Map 2 Statistics: Num rows: 1153 Data size: 129236 Basic stats: COMPLETE Column stats: NONE File Output Operator compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false Statistics: Num rows: 1153 Data size: 129236 Basic stats: COMPLETE Column stats: NONE table: input format: org.apache.hadoop.mapred.SequenceFileInputFormat output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Local Work: Map Reduce Local Work http://git-wip-us.apache.org/repos/asf/hive/blob/16d28b34/ql/src/test/results/clientpositive/spark/vector_distinct_2.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/vector_distinct_2.q.out b/ql/src/test/results/clientpositive/spark/vector_distinct_2.q.out index 803a53b..af9ec87 100644 --- a/ql/src/test/results/clientpositive/spark/vector_distinct_2.q.out +++ b/ql/src/test/results/clientpositive/spark/vector_distinct_2.q.out @@ -105,12 +105,16 @@ POSTHOOK: Lineage: vectortab2korc.si SIMPLE [(vectortab2k)vectortab2k.FieldSchem POSTHOOK: Lineage: vectortab2korc.t SIMPLE [(vectortab2k)vectortab2k.FieldSchema(name:t, type:tinyint, comment:null), ] POSTHOOK: Lineage: vectortab2korc.ts SIMPLE [(vectortab2k)vectortab2k.FieldSchema(name:ts, type:timestamp, comment:null), ] POSTHOOK: Lineage: vectortab2korc.ts2 SIMPLE [(vectortab2k)vectortab2k.FieldSchema(name:ts2, type:timestamp, comment:null), ] -PREHOOK: query: explain +PREHOOK: query: explain vectorization expression select distinct s, t from vectortab2korc PREHOOK: type: QUERY -POSTHOOK: query: explain +POSTHOOK: query: explain vectorization expression select distinct s, t from vectortab2korc POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: Stage-1 is a root stage Stage-0 depends on stages: Stage-1 @@ -127,11 +131,24 @@ STAGE PLANS: TableScan alias: vectortab2korc Statistics: Num rows: 2000 Data size: 918712 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12] Select Operator expressions: t (type: tinyint), s (type: string) outputColumnNames: t, s + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0, 8] Statistics: Num rows: 2000 Data size: 918712 Basic stats: COMPLETE Column stats: NONE Group By Operator + Group By Vectorization: + className: VectorGroupByOperator + vectorOutput: true + keyExpressions: col 0, col 8 + native: false + projectedOutputColumns: [] keys: t (type: tinyint), s (type: string) mode: hash outputColumnNames: _col0, _col1 @@ -140,12 +157,38 @@ STAGE PLANS: key expressions: _col0 (type: tinyint), _col1 (type: string) sort order: ++ Map-reduce partition columns: _col0 (type: tinyint), _col1 (type: string) + Reduce Sink Vectorization: + className: VectorReduceSinkOperator + native: false + nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true, Not ACID UPDATE or DELETE IS true, No buckets IS true, No TopN IS true, No DISTINCT columns IS true, BinarySortableSerDe for keys IS true, LazyBinarySerDe for values IS true + nativeConditionsNotMet: Uniform Hash IS false Statistics: Num rows: 2000 Data size: 918712 Basic stats: COMPLETE Column stats: NONE Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reducer 2 Execution mode: vectorized + Reduce Vectorization: + enabled: true + enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true + groupByVectorOutput: true + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reduce Operator Tree: Group By Operator + Group By Vectorization: + className: VectorGroupByOperator + vectorOutput: true + keyExpressions: col 0, col 1 + native: false + projectedOutputColumns: [] keys: KEY._col0 (type: tinyint), KEY._col1 (type: string) mode: mergepartial outputColumnNames: _col0, _col1 @@ -153,9 +196,16 @@ STAGE PLANS: Select Operator expressions: _col1 (type: string), _col0 (type: tinyint) outputColumnNames: _col0, _col1 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [1, 0] Statistics: Num rows: 1000 Data size: 459356 Basic stats: COMPLETE Column stats: NONE File Output Operator compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false Statistics: Num rows: 1000 Data size: 459356 Basic stats: COMPLETE Column stats: NONE table: input format: org.apache.hadoop.mapred.SequenceFileInputFormat http://git-wip-us.apache.org/repos/asf/hive/blob/16d28b34/ql/src/test/results/clientpositive/spark/vector_elt.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/vector_elt.q.out b/ql/src/test/results/clientpositive/spark/vector_elt.q.out index bb66867..b49462a 100644 --- a/ql/src/test/results/clientpositive/spark/vector_elt.q.out +++ b/ql/src/test/results/clientpositive/spark/vector_elt.q.out @@ -1,29 +1,79 @@ -PREHOOK: query: EXPLAIN SELECT (ctinyint % 2) + 1, cstring1, cint, elt((ctinyint % 2) + 1, cstring1, cint) +PREHOOK: query: EXPLAIN VECTORIZATION EXPRESSION SELECT (ctinyint % 2) + 1, cstring1, cint, elt((ctinyint % 2) + 1, cstring1, cint) FROM alltypesorc WHERE ctinyint > 0 LIMIT 10 PREHOOK: type: QUERY -POSTHOOK: query: EXPLAIN SELECT (ctinyint % 2) + 1, cstring1, cint, elt((ctinyint % 2) + 1, cstring1, cint) +POSTHOOK: query: EXPLAIN VECTORIZATION EXPRESSION SELECT (ctinyint % 2) + 1, cstring1, cint, elt((ctinyint % 2) + 1, cstring1, cint) FROM alltypesorc WHERE ctinyint > 0 LIMIT 10 POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: - Stage-0 is a root stage + Stage-1 is a root stage + Stage-0 depends on stages: Stage-1 STAGE PLANS: + Stage: Stage-1 + Spark +#### A masked pattern was here #### + Vertices: + Map 1 + Map Operator Tree: + TableScan + alias: alltypesorc + Statistics: Num rows: 12288 Data size: 377237 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11] + Filter Operator + Filter Vectorization: + className: VectorFilterOperator + native: true + predicateExpression: FilterLongColGreaterLongScalar(col 0, val 0) -> boolean + predicate: (ctinyint > 0) (type: boolean) + Statistics: Num rows: 4096 Data size: 125745 Basic stats: COMPLETE Column stats: NONE + Select Operator + expressions: ((UDFToInteger(ctinyint) % 2) + 1) (type: int), cstring1 (type: string), cint (type: int), elt(((UDFToInteger(ctinyint) % 2) + 1), cstring1, cint) (type: string) + outputColumnNames: _col0, _col1, _col2, _col3 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [13, 6, 2, 16] + selectExpressions: LongColAddLongScalar(col 12, val 1)(children: LongColModuloLongScalar(col 0, val 2)(children: col 0) -> 12:long) -> 13:long, VectorElt(columns [14, 6, 15])(children: LongColAddLongScalar(col 12, val 1)(children: LongColModuloLongScalar(col 0, val 2)(children: col 0) -> 12:long) -> 14:long, col 6, CastLongToString(col 2) -> 15:String) -> 16:string + Statistics: Num rows: 4096 Data size: 125745 Basic stats: COMPLETE Column stats: NONE + Limit + Number of rows: 10 + Limit Vectorization: + className: VectorLimitOperator + native: true + Statistics: Num rows: 10 Data size: 300 Basic stats: COMPLETE Column stats: NONE + File Output Operator + compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false + Statistics: Num rows: 10 Data size: 300 Basic stats: COMPLETE Column stats: NONE + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true + Stage: Stage-0 Fetch Operator limit: 10 Processor Tree: - TableScan - alias: alltypesorc - Filter Operator - predicate: (ctinyint > 0) (type: boolean) - Select Operator - expressions: ((UDFToInteger(ctinyint) % 2) + 1) (type: int), cstring1 (type: string), cint (type: int), elt(((UDFToInteger(ctinyint) % 2) + 1), cstring1, cint) (type: string) - outputColumnNames: _col0, _col1, _col2, _col3 - Limit - Number of rows: 10 - ListSink + ListSink PREHOOK: query: SELECT (ctinyint % 2) + 1, cstring1, cint, elt((ctinyint % 2) + 1, cstring1, cint) FROM alltypesorc @@ -47,7 +97,7 @@ POSTHOOK: Input: default@alltypesorc 1 cvLH6Eat2yFsyy7p 528534767 cvLH6Eat2yFsyy7p 2 cvLH6Eat2yFsyy7p 528534767 528534767 1 cvLH6Eat2yFsyy7p 528534767 cvLH6Eat2yFsyy7p -PREHOOK: query: EXPLAIN +PREHOOK: query: EXPLAIN VECTORIZATION EXPRESSION SELECT elt(2, 'abc', 'defg'), elt(3, 'aa', 'bb', 'cc', 'dd', 'ee', 'ff', 'gg'), elt('1', 'abc', 'defg'), @@ -60,7 +110,7 @@ SELECT elt(2, 'abc', 'defg'), elt(3, 'abc', 'defg') FROM alltypesorc LIMIT 1 PREHOOK: type: QUERY -POSTHOOK: query: EXPLAIN +POSTHOOK: query: EXPLAIN VECTORIZATION EXPRESSION SELECT elt(2, 'abc', 'defg'), elt(3, 'aa', 'bb', 'cc', 'dd', 'ee', 'ff', 'gg'), elt('1', 'abc', 'defg'), @@ -73,22 +123,67 @@ SELECT elt(2, 'abc', 'defg'), elt(3, 'abc', 'defg') FROM alltypesorc LIMIT 1 POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: - Stage-0 is a root stage + Stage-1 is a root stage + Stage-0 depends on stages: Stage-1 STAGE PLANS: + Stage: Stage-1 + Spark +#### A masked pattern was here #### + Vertices: + Map 1 + Map Operator Tree: + TableScan + alias: alltypesorc + Statistics: Num rows: 12288 Data size: 377237 Basic stats: COMPLETE Column stats: COMPLETE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11] + Select Operator + expressions: 'defg' (type: string), 'cc' (type: string), 'abc' (type: string), '2' (type: string), '12345' (type: string), '123456789012' (type: string), '1.25' (type: string), '16.0' (type: string), null (type: string), null (type: string) + outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [12, 13, 14, 15, 16, 17, 18, 19, 20, 21] + selectExpressions: ConstantVectorExpression(val defg) -> 12:string, ConstantVectorExpression(val cc) -> 13:string, ConstantVectorExpression(val abc) -> 14:string, ConstantVectorExpression(val 2) -> 15:string, ConstantVectorExpression(val 12345) -> 16:string, ConstantVectorExpression(val 123456789012) -> 17:string, ConstantVectorExpression(val 1.25) -> 18:string, ConstantVectorExpression(val 16.0) -> 19:string, ConstantVectorExpression(val null) -> 20:string, ConstantVectorExpression(val null) -> 21:string + Statistics: Num rows: 12288 Data size: 8687784 Basic stats: COMPLETE Column stats: COMPLETE + Limit + Number of rows: 1 + Limit Vectorization: + className: VectorLimitOperator + native: true + Statistics: Num rows: 1 Data size: 875 Basic stats: COMPLETE Column stats: COMPLETE + File Output Operator + compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false + Statistics: Num rows: 1 Data size: 875 Basic stats: COMPLETE Column stats: COMPLETE + table: + input format: org.apache.hadoop.mapred.SequenceFileInputFormat + output format: org.apache.hadoop.hive.ql.io.HiveSequenceFileOutputFormat + serde: org.apache.hadoop.hive.serde2.lazy.LazySimpleSerDe + Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true + Stage: Stage-0 Fetch Operator limit: 1 Processor Tree: - TableScan - alias: alltypesorc - Select Operator - expressions: 'defg' (type: string), 'cc' (type: string), 'abc' (type: string), '2' (type: string), '12345' (type: string), '123456789012' (type: string), '1.25' (type: string), '16.0' (type: string), null (type: string), null (type: string) - outputColumnNames: _col0, _col1, _col2, _col3, _col4, _col5, _col6, _col7, _col8, _col9 - Limit - Number of rows: 1 - ListSink + ListSink PREHOOK: query: SELECT elt(2, 'abc', 'defg'), elt(3, 'aa', 'bb', 'cc', 'dd', 'ee', 'ff', 'gg'), http://git-wip-us.apache.org/repos/asf/hive/blob/16d28b34/ql/src/test/results/clientpositive/spark/vector_groupby_3.q.out ---------------------------------------------------------------------- diff --git a/ql/src/test/results/clientpositive/spark/vector_groupby_3.q.out b/ql/src/test/results/clientpositive/spark/vector_groupby_3.q.out index e13c311..a99dfd8 100644 --- a/ql/src/test/results/clientpositive/spark/vector_groupby_3.q.out +++ b/ql/src/test/results/clientpositive/spark/vector_groupby_3.q.out @@ -105,12 +105,16 @@ POSTHOOK: Lineage: vectortab2korc.si SIMPLE [(vectortab2k)vectortab2k.FieldSchem POSTHOOK: Lineage: vectortab2korc.t SIMPLE [(vectortab2k)vectortab2k.FieldSchema(name:t, type:tinyint, comment:null), ] POSTHOOK: Lineage: vectortab2korc.ts SIMPLE [(vectortab2k)vectortab2k.FieldSchema(name:ts, type:timestamp, comment:null), ] POSTHOOK: Lineage: vectortab2korc.ts2 SIMPLE [(vectortab2k)vectortab2k.FieldSchema(name:ts2, type:timestamp, comment:null), ] -PREHOOK: query: explain +PREHOOK: query: explain vectorization expression select s, t, max(b) from vectortab2korc group by s, t PREHOOK: type: QUERY -POSTHOOK: query: explain +POSTHOOK: query: explain vectorization expression select s, t, max(b) from vectortab2korc group by s, t POSTHOOK: type: QUERY +PLAN VECTORIZATION: + enabled: true + enabledConditionsMet: [hive.vectorized.execution.enabled IS true] + STAGE DEPENDENCIES: Stage-1 is a root stage Stage-0 depends on stages: Stage-1 @@ -127,12 +131,26 @@ STAGE PLANS: TableScan alias: vectortab2korc Statistics: Num rows: 2000 Data size: 918712 Basic stats: COMPLETE Column stats: NONE + TableScan Vectorization: + native: true + projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12] Select Operator expressions: t (type: tinyint), s (type: string), b (type: bigint) outputColumnNames: t, s, b + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [0, 8, 3] Statistics: Num rows: 2000 Data size: 918712 Basic stats: COMPLETE Column stats: NONE Group By Operator aggregations: max(b) + Group By Vectorization: + aggregators: VectorUDAFMaxLong(col 3) -> bigint + className: VectorGroupByOperator + vectorOutput: true + keyExpressions: col 0, col 8 + native: false + projectedOutputColumns: [0] keys: t (type: tinyint), s (type: string) mode: hash outputColumnNames: _col0, _col1, _col2 @@ -141,14 +159,41 @@ STAGE PLANS: key expressions: _col0 (type: tinyint), _col1 (type: string) sort order: ++ Map-reduce partition columns: _col0 (type: tinyint), _col1 (type: string) + Reduce Sink Vectorization: + className: VectorReduceSinkOperator + native: false + nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true, Not ACID UPDATE or DELETE IS true, No buckets IS true, No TopN IS true, No DISTINCT columns IS true, BinarySortableSerDe for keys IS true, LazyBinarySerDe for values IS true + nativeConditionsNotMet: Uniform Hash IS false Statistics: Num rows: 2000 Data size: 918712 Basic stats: COMPLETE Column stats: NONE value expressions: _col2 (type: bigint) Execution mode: vectorized + Map Vectorization: + enabled: true + enabledConditionsMet: hive.vectorized.use.vectorized.input.format IS true + groupByVectorOutput: true + inputFileFormats: org.apache.hadoop.hive.ql.io.orc.OrcInputFormat + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reducer 2 Execution mode: vectorized + Reduce Vectorization: + enabled: true + enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true, hive.execution.engine spark IN [tez, spark] IS true + groupByVectorOutput: true + allNative: false + usesVectorUDFAdaptor: false + vectorized: true Reduce Operator Tree: Group By Operator aggregations: max(VALUE._col0) + Group By Vectorization: + aggregators: VectorUDAFMaxLong(col 2) -> bigint + className: VectorGroupByOperator + vectorOutput: true + keyExpressions: col 0, col 1 + native: false + projectedOutputColumns: [0] keys: KEY._col0 (type: tinyint), KEY._col1 (type: string) mode: mergepartial outputColumnNames: _col0, _col1, _col2 @@ -156,9 +201,16 @@ STAGE PLANS: Select Operator expressions: _col1 (type: string), _col0 (type: tinyint), _col2 (type: bigint) outputColumnNames: _col0, _col1, _col2 + Select Vectorization: + className: VectorSelectOperator + native: true + projectedOutputColumns: [1, 0, 2] Statistics: Num rows: 1000 Data size: 459356 Basic stats: COMPLETE Column stats: NONE File Output Operator compressed: false + File Sink Vectorization: + className: VectorFileSinkOperator + native: false Statistics: Num rows: 1000 Data size: 459356 Basic stats: COMPLETE Column stats: NONE table: input format: org.apache.hadoop.mapred.SequenceFileInputFormat