Hive開啟向量化模式也是hiveSQL優化方法中的一種,可以提升hive查詢速率,也叫hive矢量化,
問題1:那么什么是hive向量化模式呢?
問題2:hive向量化什么情況下可以被使用,或者說它有哪些使用場景呢?
問題3:如何查看hive向量化使用的相關資訊?
1.什么是hive向量化模式
hive向量化模式是hive的一個特性,也叫hive矢量化,在沒有引入向量化的執行模式之前,一般的查詢操作一次只處理一行資料,在向量化查詢執行時一次處理1024行的塊來簡化系統底層的操作,提高了資料處理的性能,
在底層,hive提供的向量模式,并不是重寫了Mapper函式,而是通過實作inputformat介面,創建了VectorizedParquetInputFormat類,來構建一個批量輸入的陣列,
向量化模式開啟的方式如下:
-- 開啟hive向量化模式
set hive.vectorized.execution.enabled = true;
2.Hive向量化模式支持的使用場景
Hive向量化模式并不是可以直接使用,它對使用的計算引擎,使用資料的資料型別,以及使用的SQL函式都有一定的要求,
2.1 hive向量化模式使用前置條件
-
不同的計算引擎支持程度不一樣:MR計算引擎僅支持Map階段的向量化,Tez和Spark計算引擎可以支持Map階段和Reduce階段的向量化,
-
hive檔案存盤型別必須為ORC或者Parquet等列存盤檔案型別,
2.2 向量模式支持的資料型別
- tinyint
- smallint
- int
- bigint
- boolean
- float
- double
- decimal
- date
- timestamp
- string
以上資料型別為向量化模式支持的資料型別,如果使用其他資料型別,例如array和map等,開啟了向量化模式查詢,查詢操作將使用標準的方式單行執行,但不會報錯,
2.3 向量化模式支持的函式
算數運算式: +, -, *, /, %
邏輯關系:AND, OR, NOT
比較關系(過濾器): <, >, <=, >=, =, !=, BETWEEN, IN ( list-of-constants ) as filters
使用 AND, OR, NOT, <, >, <=, >=, =, != 等布林值運算式(非過濾器)
空值校驗:IS [NOT] NULL
所有的數學函式,例如 SIN, LOG等
字串函式: SUBSTR, CONCAT, TRIM, LTRIM, RTRIM, LOWER, UPPER, LENGTH
型別轉換:cast
Hive UDF函式, 包括標準和通用的UDF函式
日期函式:YEAR, MONTH, DAY, HOUR, MINUTE, SECOND, UNIX_TIMESTAMP
IF條件運算式
以上函式運算式在運行時支持使用向量化模式,
3.如何查看hiveSQL向量化運行資訊
查看hive向量化資訊是前置的,可以通過執行計劃命令explain vectorization查看向量化描述資訊,當然,執行中,也可以通過日志了解向量化執行資訊,但相對篩選關鍵資訊比較復雜,
explain vectorization是在hive2.3.0版本之后發布的功能,可以查看map階段和reduce階段為什么沒有開啟矢量化模式,類似除錯功能,
explain vectorization支持的語法:explain vectorization [only] [summary|operator|expression|detail]
- explain vectorization:不帶后置引數,顯示執行計劃的向量化資訊(啟用向量化)以及 Map 和 Reduce 階段的摘要,
- only:這個命令只顯示向量化模式相關的描述資訊,這個引數和后面的其他引數是可以一起使用的,與它相對的是explain vectorization,
- summary:這是個默認引數,任何命令后面默認有該引數,
- operator:補充顯示運算子的向量化資訊,例如資料過濾向量化,還包括summary的所有資訊,
- expression:補充顯示運算式的向量化資訊,例如謂詞運算式,還包括 summary 和 operator 的所有資訊,
- detail:顯示最詳細級別的向量化資訊,它包括summary、operator、expression的所有資訊,
接下來我們通過實體來查看以上命令的展示內容:
3.1 explain vectorization only只查詢向量化描述資訊內容
例1 關閉向量化模式的情況下,使用explain vectorization only,
-- 關閉向量化模式
set hive.vectorized.execution.enabled = false;
explain vectorization only
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
執行結果:
PLAN VECTORIZATION:
enabled: false #標識向量化模式沒有開啟
enabledConditionsNotMet: [hive.vectorized.execution.enabled IS false] #未開啟原因
如上,如果關閉向量化模式,輸出結果中PLAN VECTORIZATION 這里可以看到該模式沒有被開啟,原因是沒有滿足enabledConditionsNotMet 指代的條件,
例2 開啟向量化模式的情況下,使用explain vectorization only,
-- 開啟向量化模式
set hive.vectorized.execution.enabled = true;
explain vectorization only
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
執行結果:
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
STAGE PLANS:
Stage: Stage-1
Map Reduce
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.parquet.MapredParquetInputFormat
allNative: false
usesVectorUDFAdaptor: false
vectorized: true
Reduce Vectorization:
enabled: false
enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true
enableConditionsNotMet: hive.execution.engine mr IN [tez, spark] IS false
Stage: Stage-0
Fetch Operator
執行結果有三部分內容:
- PLAN VECTORIZATION
- STAGE DEPENDENCIES
- STAGE PLANS
其中STAGE PLANS列印的并不是explain中map和reduce階段的運行資訊,而是這兩個階段使用向量化模式的資訊,
對以上案例內容進行關鍵詞解讀:
- Execution mode:當前的執行模式,vectorized表示當前模式是向量化的模式,
- Map Vectorization:當前是map階段的向量化執行模式資訊,
- enabled:是否開啟該階段向量化模式,true表示開啟,false表示關閉,在上面案例中Map Vectorization階段是開啟,Reduce Vectorization階段是關閉,
- enabledConditionsMet:表示當前階段,開啟向量化模式已經滿足的條件,
- enableConditionsNotMet:表示當前階段,開啟向量化模式未滿足的條件,
- groupByVectorOutput:標識該階段分組聚合操作是否開啟向量化模式,
- inputFileFormats:當前階段,輸入的檔案格式,
- allNative:是否都是本地化操作,false表示不是,
- usesVectorUDFAdaptor:值為true時,表示至少有一個向量化運算式在使用VectorUDFAdaptor(向量化udf配接器)
- vectorized:向量化模式執行是否成功,true為是向量化執行,false為不是向量化執行,
- Reduce Vectorization:reduce階段向量化模式執行資訊,
以上整個程序在map階段執行了向量化模式,在reduce階段沒有執行向量化模式,是因為上文提到的reduce階段mr計算引擎不支持,需要tez或spark計算引擎,
3.2 explain vectorization 查看hive向量化模式執行資訊
可以執行以下命令:
-- 開啟向量化模式
set hive.vectorized.execution.enabled = true;
explain vectorization only summary
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
會發現explain vectorization only
命令和explain vectorization only summary
命令執行結果完全一致,
后續其他命令也類似,explain vectorization
等同于explain vectorization summary
,summary引數是一個默認引數,可以忽略,
例3 使用explain vectorization命令查看hive向量化模式執行資訊,
-- 開啟向量化模式
set hive.vectorized.execution.enabled = true;
explain vectorization
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
其執行結果是explain和explain vectorization only兩者相加執行結果:
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
STAGE PLANS:
Stage: Stage-1
Map Reduce
Map Operator Tree:
TableScan
alias: user_info_all
Statistics: Num rows: 32634295 Data size: 783223080 Basic stats: COMPLETE Column stats: NONE
Filter Operator
predicate: ((age < 30) and (nick like '%小%')) (type: boolean)
Statistics: Num rows: 5439049 Data size: 130537176 Basic stats: COMPLETE Column stats: NONE
Select Operator ... #省略部分
# 向量化模式描述資訊
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.parquet.MapredParquetInputFormat
allNative: false
usesVectorUDFAdaptor: false
vectorized: true
Reduce Vectorization:
enabled: false
enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true
enableConditionsNotMet: hive.execution.engine mr IN [tez, spark] IS false
Reduce Operator Tree:
Group By Operator
aggregations: count(VALUE._col0)
... #省略部分
Stage: Stage-0
Fetch Operator
limit: -1
Processor Tree:
ListSink
... 為省略了一部分資訊,
3.3 使用operator查看運算子的向量化資訊
使用explain vectorization operator可以查看顯示執行計劃程序中運算子的向量化資訊和explain運行階段資訊,
簡化版為explain vectorization only operator,加only相對前者少的部分為explain運行階段資訊,下同,explain運行階段資訊我們就不查詢了,感興趣小伙伴可以自行查詢查看,
例4 簡化版為explain vectorization only operator查看hiveSQL矢量化描述資訊,
set hive.vectorized.execution.enabled = true;
explain vectorization only operator
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
執行結果:
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
STAGE PLANS:
Stage: Stage-1
Map Reduce
Map Operator Tree:
# 表掃描的向量化資訊
TableScan Vectorization:
# 讀表采用本地的向量化模式掃描
native: true
# 過濾操作的向量化資訊
Filter Vectorization:
# 過濾操作的類
className: VectorFilterOperator
# 過濾采用本地的向量化模式
native: true
# 列篩選的向量化資訊
Select Vectorization:
className: VectorSelectOperator
native: true
# 聚合操作的向量化資訊
Group By Vectorization:
className: VectorGroupByOperator
# 輸出采用向量化輸出
vectorOutput: true
#非本地操作
native: false
# reduce output向量化資訊
Reduce Sink Vectorization:
className: VectorReduceSinkOperator
native: false
# 已滿足的Reduce Sink向量化條件
nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled 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
# 不滿足的Reduce Sink向量化條件
nativeConditionsNotMet: hive.execution.engine mr IN [tez, spark] IS false, Uniform Hash IS false
# 向量化描述資訊,同explain vectorization only,不作標注了,
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.parquet.MapredParquetInputFormat
allNative: false
usesVectorUDFAdaptor: false
vectorized: true
Reduce Vectorization:
enabled: false
enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true
enableConditionsNotMet: hive.execution.engine mr IN [tez, spark] IS false
Reduce Operator Tree:
Group By Vectorization:
vectorOutput: false
native: false
Stage: Stage-0
Fetch Operator
以上內容關鍵詞在代碼塊有行注釋標注,可以看到explain vectorization only operator命令多了在explain執行計劃程序中增加了具體每一個運算子(operator)步驟的是否向量化及具體資訊,如果不滿足向量化步驟,哪些條件滿足,哪些條件不滿足,也做了標注,
3.4 使用expression顯示欄位粒度的向量化資訊
expression:補充顯示運算式的向量化資訊,例如謂詞運算式,還包括 summary 和 operator 的所有資訊,
例5 簡化版explain vectorization only expression命令查看hiveSQL執行計劃運算式的向量化資訊,
set hive.vectorized.execution.enabled = true;
explain vectorization only expression
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
執行結果:
# 同explain vectorization
PLAN VECTORIZATION:
enabled: true
enabledConditionsMet: [hive.vectorized.execution.enabled IS true]
# 同explain vectorization
STAGE DEPENDENCIES:
Stage-1 is a root stage
Stage-0 depends on stages: Stage-1
STAGE PLANS:
Stage: Stage-1
Map Reduce
Map Operator Tree:
TableScan Vectorization:
native: true
# 表示表掃描后有25列,
projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]
Filter Vectorization:
className: VectorFilterOperator
native: true
# 表示謂詞過濾少選有兩列,以及過濾條件的內容,
predicateExpression: FilterExprAndExpr(children: FilterLongColLessLongScalar(col 11, val 30), FilterStringColLikeStringScalar(col 7, pattern %小%))
Select Vectorization:
className: VectorSelectOperator
native: true
# 表示進行列篩選的具體列,這里是第12列,陣列下標為11.如果為空[],則表示任何一個列,
projectedOutputColumns: [11]
Group By Vectorization:
# 表示使用VectorUDAFCount的方法進行count計數統計以及輸出型別,
aggregators: VectorUDAFCount(ConstantVectorExpression(val 0) -> 25:int) -> bigint
className: VectorGroupByOperator
vectorOutput: true
# 聚合列
keyExpressions: col 11
native: false
# 輸出為一個新的陣列,只有一列
projectedOutputColumns: [0]
Reduce Sink Vectorization:
className: VectorReduceSinkOperator
native: false
nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled 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: hive.execution.engine mr IN [tez, spark] IS false, Uniform Hash IS false
# 向量化描述資訊,同explain vectorization only,不作標注了,
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.parquet.MapredParquetInputFormat
allNative: false
usesVectorUDFAdaptor: false
vectorized: true
Reduce Vectorization:
enabled: false
enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true
enableConditionsNotMet: hive.execution.engine mr IN [tez, spark] IS false
Reduce Operator Tree:
Group By Vectorization:
vectorOutput: false
native: false
projectedOutputColumns: null
Stage: Stage-0
Fetch Operator
以上列印資訊內容可以看出 explain vectorization only expression命令相對列印的資訊是更細粒度到欄位級別的資訊了,基本上將操作的每一列是否使用向量化處理都列印了出來,這樣我們可以很好的判斷哪些欄位型別是不支持向量化模式的,
3.5 使用detail查看最詳細級別的向量化資訊
explain vectorization only detail 查看最詳細級別的向量化資訊,它包括summary、operator、expression的所有資訊,
例6 explain vectorization only detail 查看最詳細級別的向量化資訊,
set hive.vectorized.execution.enabled = true;
explain vectorization only detail
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
執行結果:
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
# 同explain vectorization only expression
STAGE PLANS:
Stage: Stage-1
Map Reduce
Map Operator Tree:
TableScan Vectorization:
native: true
projectedOutputColumns: [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]
Filter Vectorization:
className: VectorFilterOperator
native: true
predicateExpression: FilterExprAndExpr(children: FilterLongColLessLongScalar(col 11, val 30), FilterStringColLikeStringScalar(col 7, pattern %小%))
Select Vectorization:
className: VectorSelectOperator
native: true
projectedOutputColumns: [11]
Group By Vectorization:
aggregators: VectorUDAFCount(ConstantVectorExpression(val 0) -> 25:int) -> bigint
className: VectorGroupByOperator
vectorOutput: true
keyExpressions: col 11
native: false
projectedOutputColumns: [0]
Reduce Sink Vectorization:
className: VectorReduceSinkOperator
native: false
nativeConditionsMet: hive.vectorized.execution.reducesink.new.enabled 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: hive.execution.engine mr IN [tez, spark] IS false, Uniform Hash IS false
# 向量化描述資訊這里做了更詳細的描述
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.parquet.MapredParquetInputFormat
allNative: false
usesVectorUDFAdaptor: false
vectorized: true
rowBatchContext:
dataColumnCount: 24
includeColumns: [7, 11]
dataColumns: uid:bigint, reg_time:string, cc:string, client:string, if_new:int, last_location:string, platform_reg:string, nick:string, gender:int, birthday:string, constellation:string, age:bigint, description:string, is_realname:int, realname_date:string, last_active_day:string, is_active:int, user_status:int, user_ua:string, vst_cnt:bigint, vst_dur:bigint, is_vip:int, chat_uv:bigint, chat_cnt:bigint
partitionColumnCount: 1
partitionColumns: ymd:string
scratchColumnTypeNames: bigint
Reduce Vectorization:
enabled: false
enableConditionsMet: hive.vectorized.execution.reduce.enabled IS true
enableConditionsNotMet: hive.execution.engine mr IN [tez, spark] IS false
Reduce Operator Tree:
Group By Vectorization:
vectorOutput: false
native: false
projectedOutputColumns: null
Stage: Stage-0
Fetch Operator
通過以上內容可以看出 explain vectorization only detail列印的資訊其中執行計劃部分內容和explain vectorization only expression粒度一致,在向量化描述資訊部分做了更細粒度的描述,到欄位級別,
以上就是hive向量化explain vectorization相關引數的使用,其命令在我們使用向量化模式中進行驗證支持的函式和資料型別逐步遞進,可以根據需要使用,
而hive向量化模式可以極大程度的優化hive執行速度,
4.hive向量化模式優化執行比對
例7 執行優化速度比對,
-- 代碼1 開啟向量化模式
set hive.vectorized.execution.enabled = true;
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
-- 代碼2 關閉向量化模式
set hive.vectorized.execution.enabled = false;
select age,count(0) as num from temp.user_info_all where ymd = '20230505'
and age < 30 and nick like '%小%'
group by age;
執行結果:
# 代碼1執行結果開啟向量化模式
MapReduce Total cumulative CPU time: 1 minutes 1 seconds 740 msec
Ended Job = job_1675664438694_13647623
MapReduce Jobs Launched:
Stage-Stage-1: Map: 6 Reduce: 5 Cumulative CPU: 61.74 sec HDFS Read: 367242142 HDFS Write: 1272 SUCCESS
Total MapReduce CPU Time Spent: 1 minutes 1 seconds 740 msec
OK
15 23
... # 省略資料
29 81849
Time taken: 41.322 seconds, Fetched: 31 row(s)
# 代碼2執行結果關閉向量化模式
MapReduce Total cumulative CPU time: 1 minutes 39 seconds 190 msec
Ended Job = job_1675664438694_13647754
MapReduce Jobs Launched:
Stage-Stage-1: Map: 6 Reduce: 5 Cumulative CPU: 99.19 sec HDFS Read: 367226626 HDFS Write: 1272 SUCCESS
Total MapReduce CPU Time Spent: 1 minutes 39 seconds 190 msec
OK
15 23
... # 省略資料
29 81849
Time taken: 50.724 seconds, Fetched: 31 row(s)
以上結果可以看出,開啟向量化模式執行結果查詢耗時減少,雖然減少的不多,但在CPU使用上少了三分之一的資源,可見開啟向量化模式不僅可以提高查詢速度,還可以節省查詢資源,
以上開啟向量化模式為mr引擎測驗結果,tez和spark還具有更優的執行表現,
下一期:Hive執行計劃之只有map階段SQL性能分析和解讀
按例,歡迎點擊此處關注我的個人公眾號,交流更多知識,
后臺回復關鍵字 hive,隨機贈送一本魯邊備注版珍藏大資料書籍,
轉載請註明出處,本文鏈接:https://www.uj5u.com/shujuku/554840.html
標籤:其他
上一篇:Kafka關鍵原理
下一篇:返回列表