hive参数优化-----亲测有效

hive查询的时候,导致服务器负载过高,load值飙升,服务器CPU是8个,按理说load不超过8,都应算ok的,但是,hive在部署完后没有调参,导致在执行过程中,load值达到了7.8以上,服务器连接出现问题,因此想到了调参,现在整理如下:

本次查询所用测试语句为:

select util_lnadw21ifj1579078492403.e2467e8ab37945a3869c6d3093d2c399,util_lnadw21ifj1579078492403.1cb0cb4bda2546f0a230d1955ff0d7a4,util_lnadw21ifj1579078492403.c690536676184df9a347c76c5c575661 from (select util_191pm71579078361630.e2467e8ab37945a3869c6d3093d2c399,util_191pm71579078361630.1cb0cb4bda2546f0a230d1955ff0d7a4,util_191pm71579078361630.c690536676184df9a347c76c5c575661,util_191pm71579078361630.f081b7ce95ef4e84b4d7282bc0c5e3f6 from (select data_oy3efaueda1579078258016.e2467e8ab37945a3869c6d3093d2c399,data_oy3efaueda1579078258016.1cb0cb4bda2546f0a230d1955ff0d7a4,data_oy3efaueda1579078258016.c690536676184df9a347c76c5c575661,data_oy3efaueda1579078258016.f081b7ce95ef4e84b4d7282bc0c5e3f6 from (select data_oy3efaueda1579078258016.name as e2467e8ab37945a3869c6d3093d2c399,data_oy3efaueda1579078258016.eud as 1cb0cb4bda2546f0a230d1955ff0d7a4,data_oy3efaueda1579078258016.salary as c690536676184df9a347c76c5c575661,data_oy3efaueda1579078258016.destination as f081b7ce95ef4e84b4d7282bc0c5e3f6 from  test01.employee data_oy3efaueda1579078258016) data_oy3efaueda1579078258016 inner join(select data_diskmvki1579078266957.name as 3218b1512ecd457594d59dda6638392c,data_diskmvki1579078266957.eud as 443f2b8ea393492689720d38fbddfed1,data_diskmvki1579078266957.destination as 92eb60bfeb9041fb910c18d1fc9612ef from  test01.employee_temp data_diskmvki1579078266957) data_diskmvki1579078266957 on data_oy3efaueda1579078258016.e2467e8ab37945a3869c6d3093d2c399=data_diskmvki1579078266957.3218b1512ecd457594d59dda6638392c) util_191pm71579078361630 where util_191pm71579078361630.c690536676184df9a347c76c5c575661='45000') util_lnadw21ifj1579078492403 where util_lnadw21ifj1579078492403.1cb0cb4bda2546f0a230d1955ff0d7a4='1201';

整理hive参数如下:

================================================================================

Map Reduce数量相关 
数据分片大小 (分片的数量决定map的数量) 
计算公式: splitSize = Math.max(minSize, Math.min(maxSize, blockSize)) 
set mapreduce.input.fileinputformat.split.maxsize=750000000;

单个reduce处理的数据量 (影响reduce的数量) 
计算公式: Max(1, Min(hive.exec.reducers.max [1099], ReducerStage estimate/hive.exec.reducers.bytes.per.reducer)) x hive.tez.max.partition.factor 
set hive.exec.reducers.bytes.per.reducer=629145600;

tez将会根据vertice的输出大小动态预估调整reduce的个数 
set hive.tez.auto.reducer.parallelism = true;

hive执行引擎 mr/tez/spark 
set hive.execution.engine=mr;

调整Join顺序,让多次Join产生的中间数据尽可能小,选择不同的Join策略 
set hive.cbo.enable=true;

如果数据已经根据相同的key做好聚合,那么去除掉多余的map/reduce作业 
set hive.optimize.reducededuplication=true;

如果一个简单查询只包括一个group by和order by,此处可以设置为1或2 
set hive.optimize.reducededuplication.min.reducer=4;

Map Join优化, 不太大的表直接通过map过程做join 
set hive.auto.convert.join=true; 
set hive.auto.convert.join.noconditionaltask=true;

Map Join任务HashMap中key对应value数量 
set hive.smbjoin.cache.rows=10000;

可以被转化为HashMap放入内存的表的大小(官方推荐853M) 
set hive.auto.convert.join.noconditionaltask.size=894435328;

**map端聚合(跟group by有关), 如果开启, Hive将会在map端做第一级的聚合, 会用更多的内存 
http://dmtolp**eko.com/2014/10/13/map-side-aggregation-in-hive/ 
开启这个参数 sum(1)会有类型转换问题 
set hive.map.aggr=false;

所有map任务可以用作Hashtable的内存百分比, 如果OOM, 调小这个参数(官方默认0.5) 
set hive.map.aggr.hash.percentmemory=0.5;

将只有SELECT, FILTER, LIMIT转化为FETCH, 减少等待时间 
set hive.fetch.task.conversion=more; 
set hive.fetch.task.conversion.threshold=1073741824;

聚合查询是否转化为FETCH 
set hive.fetch.task.aggr=false;

如果数据按照join的key分桶,hive将简单优化inner join(官方推荐关闭) 
set hive.optimize.bucketmapjoin= false; 
set hive.optimize.bucketmapjoin.sortedmerge=false;

以下两个参数用于开启动态分区 
set hive.exec.dynamic.partition=true; 
set hive.exec.dynamic.partition.mode=nonstrict;

合并小文件 
set hive.merge.mapfiles=true; 
set hive.merge.mapredfiles=true; 
set hive.merge.tezfiles=true; 
set hive.merge.sparkfiles=false; 
set hive.merge.size.per.task=536870912; 
set hive.merge.smallfiles.avgsize=536870912; 
set hive.merge.orcfile.stripe.level=true;

如果开启将会在ORC文件中记录metadata 
set hive.orc.splits.include.file.footer=false;

ORC写缓冲大小 
set hive.exec.orc.default.stripe.size=67108864;

新创建的表/分区是否自动计算统计数据 
set hive.stats.autogather=true; 
set hive.compute.query.using.stats=true; 
set hive.stats.fetch.column.stats=true; 
set hive.stats.fetch.partition.stats=true;

手动统计已经存在的表 (本次没做)
ANALYZE TABLE COMPUTE STATISTICS; 
ANALYZE TABLE COMPUTE STATISTICS for COLUMNS; 
ANALYZE TABLE partition (coll=”x”) COMPUTE STATISTICS for COLUMNS;

在order by limit查询中分配给存储Top K的内存为10% 
set hive.limit.pushdown.memory.usage=0.1;

是否开启自动使用索引 
set hive.optimize.index.filter=true;

获取文件块路径的工作线程数 
set mapreduce.input.fileinputformat.list-status.num-threads=5;

如果自动分区数大于这个参数,将会报错 
set hive.exec.max.dynamic.partitions=100000; 
set hive.exec.max.dynamic.partitions.pernode=100000;

=========================================================================

对比了上述参数,本次优化了以下内容:

hive> set hive.limit.pushdown.memory.usage;
hive.limit.pushdown.memory.usage=-1.0

hive> set hive.stats.fetch.column.stats;
hive.stats.fetch.column.stats=false

hive> set hive.merge.smallfiles.avgsize;
hive.merge.smallfiles.avgsize=16000000

hive> set hive.merge.tezfiles;
hive.merge.tezfiles=false

hive> set hive.exec.dynamic.partition.mode;
hive.exec.dynamic.partition.mode=strict

hive> set hive.auto.convert.join.noconditionaltask.size;
hive.auto.convert.join.noconditionaltask.size=10000000

hive> set hive.optimize.index.filter;
hive.optimize.index.filter=false

hive> set mapreduce.input.fileinputformat.list-status.num-threads;
mapreduce.input.fileinputformat.list-status.num-threads=1

hive> set hive.exec.max.dynamic.partitions;
hive.exec.max.dynamic.partitions=1000

hive> set hive.exec.max.dynamic.partitions.pernode;
hive.exec.max.dynamic.partitions.pernode=100

hive> set hive.map.aggr;
hive.map.aggr=true

这些都是hive默认的值,按照上述均作出了调整,目前和上述参数设置一致,再次执行查询语句,load稳定在5.5左右,如果让开发单独执行这条语句,load在2以下,目前看来参数是有效的

mapred.child.java.opts 这个参数我这里没有具体优化,这个参数是配置每个map或reduce使用的内存数量。默认的是200M。对于这个参数,我个人认为,如果内存是8G,CPU有8个核,那么就设置成1G就可以了。实际上,在map和reduce的过程中对内存的消耗并不大,但是如果配置的太小,则有可能出现”无可分配内存”的错误。所以,对于这个配置我总结了一个简单的公式:map/reduce的并发数量(总和不大于CPU核数)×mapred.child.java.opts < 该节点机器的总内存。当然也可以等于,不过有点风险而已。参考:https://blog.csdn.net/breakout_alex/article/details/89019803

 

本次优化参考链接:

https://blog.csdn.net/yu0_zhang0/article/details/81776459

https://blog.csdn.net/scgaliguodong123_/article/details/45477323

https://www.cnblogs.com/felixzh/p/8604188.html

https://blog.csdn.net/young_0609/article/details/84593316

这是精髓:

https://blog.csdn.net/qq_26442553/article/details/99438121

https://blog.csdn.net/qq_26442553/article/details/99693490

https://blog.csdn.net/qq_18838991/article/details/51819295

另外:MR的内存优化参考这篇:

https://blog.csdn.net/u014665013/article/details/80923044

https://www.jianshu.com/p/73d9ce671261

hive的存储,参考这篇:

https://blog.csdn.net/zyzzxycj/article/details/79267635

https://blog.csdn.net/zyzzxycj/article/details/79270051

小文件处理,参考这篇:

https://blog.csdn.net/u010010664/article/details/83105174

 

你可能感兴趣的:(hive,hive,hive参数调优,hive优化)