You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kylin.apache.org by vishalchm <ch...@gmail.com> on 2021/03/03 13:10:15 UTC
"Extract Fact Table Distinct Columns" step runs longer
hi Folks,
i am trying to build a cube on EMR r5dxlarge * 4 nodes on a flat fact table
which has 19 dimensions and 5 measures. In process of cube building "Extract
Fact Table Distinct Columns" step takes too long. can you tell me if i am
missing anything here? could it be due to the lower grain data?
Thanks,
Vishal
--
Sent from: http://apache-kylin.74782.x6.nabble.com/
Re: "Extract Fact Table Distinct Columns" step runs longer
Posted by elkan1788 <el...@gmail.com>.
Hi, vishalchm:
From your reply that could be judged it maybe your resource were busy when
you first time running this job. Please check it detail when next time meet
the same problems. By the way, give you some tips, you can use the yarn
command pull all logs from mapreduce job. Like this:
yarn logs -applicationId "which job you see the mapreduce running" >
/tmp/yarn.log
Good luck for your travel in Kylin.
--
Sent from: http://apache-kylin.74782.x6.nabble.com/
Re: "Extract Fact Table Distinct Columns" step runs longer
Posted by vishalchm <ch...@gmail.com>.
hi,
my observation. so this process runs slow for the first time but when i
build cube next time it is completing in few minutes only. shareing the log
from second step.. looks like i am missing some mapreduce configuration here
1st attempt:
Counters: 52
File System Counters
FILE: Number of bytes read=1404339530
FILE: Number of bytes written=2100591948
FILE: Number of read operations=0
FILE: Number of large read operations=0
FILE: Number of write operations=0
HDFS: Number of bytes read=9312271
HDFS: Number of bytes written=520516128
HDFS: Number of read operations=59
HDFS: Number of large read operations=0
HDFS: Number of write operations=50
Job Counters
Killed map tasks=1
Killed reduce tasks=2
Launched map tasks=3
Launched reduce tasks=18
Data-local map tasks=3
Total time spent by all maps in occupied slots (ms)=357469920
Total time spent by all reduces in occupied slots (ms)=3550570080
Total time spent by all map tasks (ms)=1489458
Total time spent by all reduce tasks (ms)=7397021
Total vcore-milliseconds taken by all map tasks=1489458
Total vcore-milliseconds taken by all reduce tasks=7397021
Total megabyte-milliseconds taken by all map tasks=11439037440
Total megabyte-milliseconds taken by all reduce tasks=113618242560
Map-Reduce Framework
Map input records=368816
Map output records=104656
Map output bytes=1021571917
Map output materialized bytes=698471609
Input split bytes=6220
Combine input records=209310
Combine output records=209308
Reduce input groups=59717
Reduce shuffle bytes=698471609
Reduce input records=104654
Reduce output records=0
Spilled Records=313962
Shuffled Maps =34
Failed Shuffles=0
Merged Map outputs=34
GC time elapsed (ms)=5942
CPU time spent (ms)=1203260
Physical memory (bytes) snapshot=18783621120
Virtual memory (bytes) snapshot=279152558080
Total committed heap usage (bytes)=43746066432
Shuffle Errors
BAD_ID=0
CONNECTION=0
IO_ERROR=0
WRONG_LENGTH=0
WRONG_MAP=0
WRONG_REDUCE=0
File Input Format Counters
Bytes Read=0
File Output Format Counters
Bytes Written=0
org.apache.kylin.engine.mr.steps.FactDistinctColumnsMapper$RawDataCounter
BYTES=55368910
2nd attempt:
Counters: 51
File System Counters
FILE: Number of bytes read=8209556
FILE: Number of bytes written=25262878
FILE: Number of read operations=0
FILE: Number of large read operations=0
FILE: Number of write operations=0
HDFS: Number of bytes read=9336990
HDFS: Number of bytes written=1747432
HDFS: Number of read operations=87
HDFS: Number of large read operations=0
HDFS: Number of write operations=50
Job Counters
Killed reduce tasks=1
Launched map tasks=9
Launched reduce tasks=17
Data-local map tasks=9
Total time spent by all maps in occupied slots (ms)=17504880
Total time spent by all reduces in occupied slots (ms)=36970080
Total time spent by all map tasks (ms)=72937
Total time spent by all reduce tasks (ms)=77021
Total vcore-milliseconds taken by all map tasks=72937
Total vcore-milliseconds taken by all reduce tasks=77021
Total megabyte-milliseconds taken by all map tasks=560156160
Total megabyte-milliseconds taken by all reduce tasks=1183042560
Map-Reduce Framework
Map input records=368816
Map output records=90281
Map output bytes=12469333
Map output materialized bytes=8850914
Input split bytes=27990
Combine input records=90281
Combine output records=90272
Reduce input groups=27059
Reduce shuffle bytes=8850914
Reduce input records=90272
Reduce output records=0
Spilled Records=180544
Shuffled Maps =153
Failed Shuffles=0
Merged Map outputs=153
GC time elapsed (ms)=11067
CPU time spent (ms)=207100
Physical memory (bytes) snapshot=24891486208
Virtual memory (bytes) snapshot=339687297024
Total committed heap usage (bytes)=63315640320
Shuffle Errors
BAD_ID=0
CONNECTION=0
IO_ERROR=0
WRONG_LENGTH=0
WRONG_MAP=0
WRONG_REDUCE=0
File Input Format Counters
Bytes Read=0
File Output Format Counters
Bytes Written=0
org.apache.kylin.engine.mr.steps.FactDistinctColumnsMapper$RawDataCounter
BYTES=55368910
--
Sent from: http://apache-kylin.74782.x6.nabble.com/
Re: "Extract Fact Table Distinct Columns" step runs longer
Posted by elkan1788 <el...@gmail.com>.
Hi vishalchm,
Thanks for your message. I had look some info from logs, but it no complete
that not found any help message. And there your data column feature see not
problem. What kind of aggregation group you designed, please show it. There
I find out your job is still running, if it was the problem job, can you
update this job diagnose log file?
--
Sent from: http://apache-kylin.74782.x6.nabble.com/
Re: "Extract Fact Table Distinct Columns" step runs longer
Posted by vishalchm <ch...@gmail.com>.
Attached kylin.out and Kylin.log files.
Cluster configuration: r5d.4xlarge 4 nodes
*Data Details*:
Record counts - 66 million
Dimensions - 13
Measures - 5
Data Cardinality:
*Column Data Type Cardinality*
12 decimal(38,18) 66503696
13 decimal(38,18) 800325
14 decimal(38,18) 555533
15 decimal(38,18) 349654
16 decimal(38,18) 348604
17 varchar(256) 16647
18 varchar(256) 14746
19 varchar(256) 2942
20 varchar(256) 2468
21 varchar(256) 635
22 varchar(256) 541
23 varchar(256) 382
24 varchar(256) 131
25 varchar(256) 61
26 varchar(256) 48
27 varchar(256) 8
28 varchar(256) 6
29 varchar(256) 2
30 varchar(256) 1
31 timestamp 1
32 varchar(256) 1
kylin.log <http://apache-kylin.74782.x6.nabble.com/file/t1164/kylin.log>
kylin.out <http://apache-kylin.74782.x6.nabble.com/file/t1164/kylin.out>
--
Sent from: http://apache-kylin.74782.x6.nabble.com/
Re: "Extract Fact Table Distinct Columns" step runs longer
Posted by elkan1788 <el...@gmail.com>.
Hi vishalchm,
There were so many possibilities, and I think you give message was not
enough! Please tell about which version kylin you were using, and the data
feature , just like the data cardinality etc. also could you upload some
kylin log info which is better way to figure out the root case.
--
Sent from: http://apache-kylin.74782.x6.nabble.com/
Re: "Extract Fact Table Distinct Columns" step runs longer
Posted by vishalchm <ch...@gmail.com>.
its r5d.4xlarge
--
Sent from: http://apache-kylin.74782.x6.nabble.com/