INSERT OVERWRITE TABLE result
SELECT /*+ STREAMTABLE(product) */
i.IMAGE_ID,
p.PRODUCT_NO,
p.STORE_NO,
p.PRODUCT_CAT_NO,
p.CAPTION,
p.PRODUCT_DESC,
p.IMAGE1_ID,
p.IMAGE2_ID,
s.STORE_ID,
s.STORE_NAME,
p.CREATE_DATE,
CASE WHEN custImg.IMAGE_ID is NULL THEN 0 ELSE 1 END,
CASE WHEN custImg1.IMAGE_ID is NULL THEN 0 ELSE 1 END,
CASE WHEN custImg2.IMAGE_ID is NULL THEN 0 ELSE 1 END
FROM image i
JOIN PRODUCT p ON i.IMAGE_ID = p.IMAGE1_ID
JOIN PRODUCT_CAT pcat ON p.PRODUCT_CAT_NO = pcat.PRODUCT_CAT_NO
JOIN STORE s ON p.STORE_NO = s.STORE_NO
JOIN STOCK_INFO si ON si.STOCK_INFO_ID = pcat.STOCK_INFO_ID
LEFT OUTER JOIN CUSTOMIZABLE_IMAGE custImg ON i.IMAGE_ID = custImg.IMAGE_ID
LEFT OUTER JOIN CUSTOMIZABLE_IMAGE custImg1 ON p.IMAGE1_ID = custImg1.IMAGE_ID
LEFT OUTER JOIN CUSTOMIZABLE_IMAGE custImg2 ON p.IMAGE2_ID = custImg2.IMAGE_ID;
I have a join query where i am joining huge tables and i am trying to optimize this hive query. Here are some facts about the tables
image table has 60m rows,
product table has 1b rows,
product_cat has 1000 rows,
store has 1m rows,
stock_info has 100 rows,
customizable_image has 200k rows.
a product can have one or two images (image1 and image2) and product level information are stored only in product table. i tried moving the join with product to the bottom but i couldnt as all other following joins require data from the product table.
Here is what i tried so far,
1. I gave the hint to hive to stream product table as its the biggest one
2. I bucketed the table (during create table) into 256 buckets (on image_id) and then did the join - didnt give me any significant performance gain
3. changed the input format to sequence file from textfile(gzip files) , so that it can be splittable and hence more mappers can be run if hive want to run more mappers
Here are some key logs from hive console. I ran this hive query in aws. Can anyone help me understand the primary bottleneck here ? This job is only processing a subset of the actual data.
Stage-14 is selected by condition resolver.
Launching Job 1 out of 11
Number of reduce tasks not specified. Estimated from input data size: 22
In order to change the average load for a reducer (in bytes):
set hive.exec.reducers.bytes.per.reducer=<number>
In order to limit the maximum number of reducers:
set hive.exec.reducers.max=<number>
In order to set a constant number of reducers:
set mapred.reduce.tasks=<number>
Kill Command = /home/hadoop/bin/hadoop job -kill job_201403242034_0001
Hadoop job information for Stage-14: number of mappers: 341; number of reducers: 22
2014-03-24 20:55:05,709 Stage-14 map = 0%, reduce = 0%
.
2014-03-24 23:26:32,064 Stage-14 map = 100%, reduce = 100%, Cumulative CPU 34198.12 sec
MapReduce Total cumulative CPU time: 0 days 9 hours 29 minutes 58 seconds 120 msec
.
2014-03-25 00:33:39,702 Stage-30 map = 100%, reduce = 100%, Cumulative CPU 20879.69 sec
MapReduce Total cumulative CPU time: 0 days 5 hours 47 minutes 59 seconds 690 msec
.
2014-03-26 04:15:25,809 Stage-14 map = 100%, reduce = 100%, Cumulative CPU 3903.4 sec
MapReduce Total cumulative CPU time: 0 days 1 hours 5 minutes 3 seconds 400 msec
.
2014-03-26 04:25:05,892 Stage-30 map = 100%, reduce = 100%, Cumulative CPU 2707.34 sec
MapReduce Total cumulative CPU time: 45 minutes 7 seconds 340 msec
.
2014-03-26 04:45:56,465 Stage-2 map = 100%, reduce = 100%, Cumulative CPU 3901.99 sec
MapReduce Total cumulative CPU time: 0 days 1 hours 5 minutes 1 seconds 990 msec
.
2014-03-26 04:54:56,061 Stage-26 map = 100%, reduce = 100%, Cumulative CPU 2388.71 sec
MapReduce Total cumulative CPU time: 39 minutes 48 seconds 710 msec
.
2014-03-26 05:12:35,541 Stage-4 map = 100%, reduce = 100%, Cumulative CPU 3792.5 sec
MapReduce Total cumulative CPU time: 0 days 1 hours 3 minutes 12 seconds 500 msec
.
2014-03-26 05:34:21,967 Stage-5 map = 100%, reduce = 100%, Cumulative CPU 4432.22 sec
MapReduce Total cumulative CPU time: 0 days 1 hours 13 minutes 52 seconds 220 msec
.
2014-03-26 05:54:43,928 Stage-21 map = 100%, reduce = 100%, Cumulative CPU 6052.96 sec
MapReduce Total cumulative CPU time: 0 days 1 hours 40 minutes 52 seconds 960 msec
MapReduce Jobs Launched:
Job 0: Map: 59 Reduce: 18 Cumulative CPU: 3903.4 sec HDFS Read: 37387 HDFS Write: 12658668325 SUCCESS
Job 1: Map: 48 Cumulative CPU: 2707.34 sec HDFS Read: 12658908810 HDFS Write: 9321506973 SUCCESS
Job 2: Map: 29 Reduce: 10 Cumulative CPU: 3901.99 sec HDFS Read: 9321641955 HDFS Write: 11079251576 SUCCESS
Job 3: Map: 42 Cumulative CPU: 2388.71 sec HDFS Read: 11079470178 HDFS Write: 10932264824 SUCCESS
Job 4: Map: 42 Reduce: 12 Cumulative CPU: 3792.5 sec HDFS Read: 10932405443 HDFS Write: 11812454443 SUCCESS
Job 5: Map: 45 Reduce: 13 Cumulative CPU: 4432.22 sec HDFS Read: 11812679475 HDFS Write: 11815458945 SUCCESS
Job 6: Map: 42 Cumulative CPU: 6052.96 sec HDFS Read: 11815691155 HDFS Write: 0 SUCCESS
Total MapReduce CPU Time Spent: 0 days 7 hours 32 minutes 59 seconds 120 msec
OK
The query is still taking longer than 5 hours in Hive where as in RDBMS it takes only 5 hrs. I need some help in optimizing this query, so that it executes much faster. Interestingly, when i ran the task with 4 large core instances, the time taken improved only by 10 mins compared to the run with 3 large instance core instances. but when i ran the task with 3 med cores, it took 1hr 10 mins more.
This brings me to the question, "is Hive even the right choice for such complex joins" ?
I suspect the bottleneck is just in sorting your product table, since it seems much larger than the others. I think joins with Hive for tables over a certain size become untenable, simply because they require a sort.
There are parameters to optimize sorting, like io.sort.mb, which you can try setting, so that more sorting occurs in memory, rather than spilling to disk, re-reading and re-sorting. Look at the number of spilled records, and see if this much larger than your inputs. There are a variety of ways to optimize sorting. It might also help to break your query up into multiple subqueries so it doesn't have to sort as much at one time.
For the stock_info , and product_cat tables, you could probably keep them in memory since they are so small ( Check out the 'distributed_map' UDF in Brickhouse ( https://github.com/klout/brickhouse/blob/master/src/main/java/brickhouse/udf/dcache/DistributedMapUDF.java ) For custom image, you might be able to use a bloom filter, if having a few false positives is not a real big problem.
To completely remove the join, perhaps you could store the image info in a keystone DB like HBase to do lookups instead. Brickhouse also had UDFs for HBase , like hbase_get and base_cached_get .
Related
I have a small data set of ~200 samples taken over twenty years with two columns of data that sometimes have multiple entries for the period (i.e. age or date). When I go to plot it, even though the data is over 20 years the graph heavily reflects the number of samples in the period and not the period itself. For example during age 23 there may be 2 or 3 samples, 1 for age 24, 20 for age 25, and 10 for age 35.. the number of samples entirely on needs for additional data at the time.. so simply there is no consistency to the sample rate.
How do I get an Max or an Average / Max for a period (age) and ensure there is only one entry per period in the sheet (about one entry per year) without having to create a separate sheet full of separate queries and charting off of that?
What I have tried in Google Sheets (where my data is) is on the x-series chart choosing "aggregate" (which is on the age period) which helps flatten the graph a bit, but doesn't reduce the series.
A read only link to the the spreadsheet is HERE for reference.
Data Looking something like this:
3/27/2013 36.4247 2.5 29.3
4/10/2013 36.4630 1.8 42.8
4/15/2013 36.4767 2.2 33.9
5/2/2013 36.5233 2.2 33.9
5/21/2013 36.5753 1.91 39.9
5/29/2013 36.5973 1.94 39.2
7/29/2013 36.7644 1.98 38.3
10/25/2013 37.0055 1.7 45.6
2/28/2014 37.3507 1.85 50 41.3
6/1/2014 37.6055 1.98 38 38.1
12/1/2014 38.1068 37
6/1/2015 38.6055 2.18 34 33.9
12/11/2015 39.1342 3.03 23 23.1
12/14/2015 39.1425 3.18 22 21.9
12/15/2015 39.1452 3.44 20 20.0
12/17/2015 39.1507 3.61 19 18.9
12/21/2015 39.1616 3.62 19 18.8
12/23/2015 39.1671 3.32 21 20.8
12/25/2015 39.1726 3.08 23 22.7
12/28/2015 39.1808 3.12 22 22.4
12/29/2015 39.1836 2.97 24 23.7
12/30/2015 39.1863 3.57 19 19.1
12/31/2015 39.1890 3.37 20 20.5
1/1/2016 39.1918 3.37 20 20.5
1/3/2016 39.1973 2.65 27 27.0
1/4/2016 39.2000 2.76 26 25.8
try:
=QUERY(SORTN(SORT({YEAR($A$6:$A), B6:B}, 1, 0, 2, 0), 9^9, 2, 1, 1),
"where Col1 <> 1899")
demo spreadsheet
and build a chart from there
I just running a simple insertion query in Enterprise version neo4j-3.1.3 on neo4j browser. At the very first time insertion execution time is 410ms, and subsequent insertion it reduce to 4ms.
CQL:
create (n:City {name:"Trichy", lat:50.25, lng:12.21});
//Execution Time : Completed after 410 ms.
Even fetching a single node taking much time.
My CQL Query:
MATCH (n:City) RETURN n LIMIT 25
//Execution time : Started streaming 1 record after 105 ms and completed after 111 ms
I have allotted: dbms.memory.pagecache.size=15g
System:
total used free shared buffers cached
Mem: 11121 8171 2950 611 956 2718
-/+ buffers/cache: 4496 6625
Swap: 3891 0 3891
Why its taking much time on singe insertion. Even 4ms seems too costly for a single insertion with minimal property.
Even fetching also taking much time.
I'm using Ipython parallel in an optimisation algorithm that loops a large number of times. Parallelism is invoked in the loop using the map method of a LoadBalancedView (twice), a DirectView's dictionary interface and an invocation of a %px magic. I'm running the algorithm in an Ipython notebook.
I find that the memory consumed by both the kernel running the algorithm and one of the controllers increases steadily over time, limiting the number of loops I can execute (since available memory is limited).
Using heapy, I profiled memory use after a run of about 38 thousand loops:
Partition of a set of 98385344 objects. Total size = 18016840352 bytes.
Index Count % Size % Cumulative % Kind (class / dict of class)
0 5059553 5 9269101096 51 9269101096 51 IPython.parallel.client.client.Metadata
1 19795077 20 2915510312 16 12184611408 68 list
2 24030949 24 1641114880 9 13825726288 77 str
3 5062764 5 1424092704 8 15249818992 85 dict (no owner)
4 20238219 21 971434512 5 16221253504 90 datetime.datetime
5 401177 0 426782056 2 16648035560 92 scipy.optimize.optimize.OptimizeResult
6 3 0 402654816 2 17050690376 95 collections.defaultdict
7 4359721 4 323814160 2 17374504536 96 tuple
8 8166865 8 196004760 1 17570509296 98 numpy.float64
9 5488027 6 131712648 1 17702221944 98 int
<1582 more rows. Type e.g. '_.more' to view.>
You can see that about half the memory is used by IPython.parallel.client.client.Metadata instances. A good indicator that results from the map invocations are being cached is the 401177 OptimizeResult instances, the same number as the number of optimize invocations via lbview.map - I am not caching them in my code.
Is there a way I can control this memory usage on both the kernel and the Ipython parallel controller (who'se memory consumption is comparable to the kernel)?
Ipython parallel clients and controllers store past results and other metadata from past transactions.
The IPython.parallel.Client class provides a method for clearing this data:
Client.purge_everything()
documented here. There is also purge_results() and purge_local_results() methods that give you some control over what gets purged.
I want to subtract a number form a duration but not sure how can I do it.
A1 : 137:47:00 (formatted as duration)
A2 : 126 (formatted as number)
When I subtract it is showing unexpected value
=(A1-A2) = -120.26
I was expecting something similar to 11.
Subtracting a number (without dimension) from a duration does not really make a lot of sense but if 137:47:00 represented 137 hours and 47 minutes then subtracting 126 hours from that would (and give a result between 11 and 12 hours). To be able to compare like with like, the duration can be represented as a number by accessing the fact that Google spreadsheets treats 24 hours as number 1. So multiply 137:47:00 (if representing hour, minutes, seconds) by 24 to get a number from which another number can be subtracted to give a meaningful result (ie 11.7833333 - representing 11 hours 47 minutes if to subtract 126 hours from 137 hours and 47 minutes). Therefore:
=24*A1-A2
might suit.
Calculating time worked per day on Web Applications addresses a vaguely similar issue.
I have data from an experiment that is sampling responses between 59 to 60 hz. There is no way to predict the drop-down in sampling rate throughout the experiment which runs for 18 minutes.
Each of the sampled responses are numbered from 1 to N (for total number of rows) showing relative passage of time, stored in variable 'frame'. I also have a unix time stamp marking absolute time stored in 'unixtime'. But unixtime is reported in whole integers & not in fractional units. For example:
1376925380 may be repeated 59 times;
1376925381 may be repeated 60 times in the data file.
I would like to create a new variable that tracks each consecutive frame (or sampled response) from 1 to 60 or from 1 to 59, as the case may be, for each given unixtime stamp in SPSS. See the desired re-arrangement below. Any help w/ appropriate SPSS-syntax is appreciated!
unixtime newframe
1376925380 1
1376925380 2
1376925380 3
1376925380 4
1376925380 5
1376925380 6
....
1376925380 58
1376925380 59
1376925381 1
1376925381 2
1376925381 3
1376925381 4
.... ....
1376925381 60
1376925382 1
1376925382 2
....
If I understand correctly, you can use LAG to figure out your counter between the time stamps. Example below.
*fake data.
set seed 10.
input program.
loop #i = 1 to 100.
loop #j = 1 to TRUNC(RV.UNIFORM(59,61)).
compute unixtime = 1376925379 + #i.
end case.
end loop.
end loop.
end file.
end input program.
*Using lag to calculate newframe variable.
DO IF ($casenum = 1) OR (unixtime <> lag(unixtime)).
compute newframe = 1.
ELSE.
compute newframe = lag(newframe) + 1.
END IF.
exe.
See related discussion for using lag at, Using sequential case processing for data management in SPSS.