Hadoop Hive MAX gives multiple results - hadoop

I am trying to get a maximum value from a count selecting 2 label srcip and max, but everytime I include srcip I have to use group by srcip at the end and gives me result as the max wasnt even there.
When I write the query like this it gives me the correct max value but I want to select srcip as well.
Select max(count1) as maximum
from (SELECT srcip,count(srcip) as count1 from data group by srcip)t;
But when I do include srcip in the select I get result as there was no max function
Select srcip,max(count1) as maximum
from (SELECT srcip,count(srcip) as count1 from data group by srcip)t
group by srcip;
I would expect from this a single result but I get multiple.
Anyone has any ideas?

You may do ORDER BY count DESC with LIMIT 1 to get the scrip with MAX of count.
SELECT srcip, count(srcip) as count1
from data group by srcip
ORDER BY count1 DESC LIMIT 1

let's consider you have a data like this.
Table
Let's see what happens when you run following query, what happens to data.
Query
SELECT srcip,count(srcip) as count1 from data group by srcip
Output: table1
Now let's see what happens you run your outer query on above table .
Select srcip,max(count1) as maximum from table1 group by srcip
Same Output
Reason being your query says to select srcip and maximum of count from each group of srcip. And we have 3 groups, so 3 rows.

The query below returns exact one row having the max count and the associated scrip. This is the query based on the expected result; you would rather look more into sql and earlier comments, then progress to hive analytical queries.
Some people could argue that there is better way to optimize this query for your expected result but this should give you a motivation to look more into Hive analytical queries.
select scrip, count1 as maximum from (select srcip, count(scrip) over (PARTITION by scrip) as count1, row_number() over (ORDER by scrip desc) as row_num from data) q1 having row_num = 1;

Related

Impala : Running sum of 1 hour

I want to count records of each ID with in 1 Hour. I tried out some IMPALA queries but without any luck.
I have input data as follows:
And expected output would be:
I tried :
select
concat(month,'/',day,'/',year,' ',hour,':',minute) time, id,
count(1) over(partition by id order by concat(month,'/',day,'/',year,' ',hour,':',minute) range between '1 hour' PRECEDING AND CURRENT ROW) request
from rt_request
where
concat(year,month,day,hour) >= '2019020318'
group by id, concat(month,'/',day,'/',year,' ',hour,':',minute)
But I got exception.
RANGE is only supported with both the lower and upper bounds UNBOUNDED or one UNBOUNDED and the other CURRENT ROW.
Any suggestion/help would be appreciated.
Thank you in advance!
I think you are looking for counts for the same hour across days for a given id. You can simply use row_number to do this.
select time,id,row_number() over(partition by id,hour order by concat(month,'/',day,'/',year,' ',hour,':',minute)) as total
from tbl

How to set range for limit clause in hive

How to set range for limit clause in hive , I have tried the below query but failed with syntax error . Can someone please help
select * from table limit 1000,2000;
You can use Row_Number window function and set the range limit.
Below Query will result only the first 20 records from the table
hive> select * from
(
SELECT *,ROW_NUMBER() over (Order by id) as rowid FROM <tab_name>
)t
where rowid > 0 and rowid <=20;
Using Between operator to specify range
hive> select * from
(
SELECT *,ROW_NUMBER() over (Order by id) as rowid FROM <tab_name>
)t
where rowid between 0 and 20;
To fetch rows from 20 to 40 then increase the lower/upper bound values
hive> select * from
(
SELECT *,ROW_NUMBER() over (Order by id) as rowid FROM <tab_name>
)t
where rowid > 20 and rowid <=40;
The LIMIT clause is used to set a ceiling on the number of rows in the result set. You are getting a syntax error because of an incorrect usage of this HQL clause.
The query could be written as the following to return no more than 2000 rows:
SELECT * FROM table LIMIT 2000;
You could also write it like so to return no more than 1000 rows:
SELECT * FROM table LIMIT 1000;
However you cannot combine both into the same argument for LIMIT. The LIMIT argument must evaluate to a constant value.
I will try and expand on this information a bit to try and help solve your problem. If you are attempting to "paginate" your results the following may be of use.
FIRST I would recommend against leaning on HQL for pagination, in most situations that would be more efficiently implemented on the application logic side (query large result set, cache what you need, paginate with application logic). If you have no choice but to pull out ranges of rows you can get the desired effect through a combination of the LIMIT, ORDER BY, and OFFSET clauses.
LIMIT : This will limit your result set to a maximum number of row
ORDER BY: This will sort/order your result set based on one or more columns
OFFSET: This will start your result set at a certain row after the logical first entry in the table.
You may combine these three clauses to effectively query "pages" of your table. For example the following three queries show how to get the first 3 blocks of data from a table where each block contains 1000 rows and the target table's 'column1' is used to determine logical order.
SELECT title as "Page 1", column1, column2, ... FROM table
ORDER BY column1 LIMIT 1000 OFFSET 0;
SELECT title as "Page 2", column1, column2, ... FROM table
ORDER BY column1 LIMIT 1000 OFFSET 1000;
SELECT title as "Page 3", column1, column2, ... FROM table
ORDER BY column1 LIMIT 1000 OFFSET 2000;
Each query declares 'column1' as the sorting value with ORDER BY. The queries will return no more than 1000 rows due to the LIMIT clause. Each result set will start at a different row due to the OFFSET being incremented by the "page size" for each query.
I am not sure what you are trying to achieve, but ...
That will return the 1001 and the 2001 record in the query results set only if you are using hive a hive version greater than 2.0.0
hive --version
(https://issues.apache.org/jira/browse/HIVE-11531)
Limit in Hive gives 'n' number of records randomly. It's not to print a range of records.
You may use order by in conjunction with limit to get what you want

HIVE equivalent of FIRST and LAST

I have a table with 3 columns:
table1: ID, CODE, RESULT, RESULT2, RESULT3
I have this SAS code:
data table1
set table1;
BY ID, CODE;
IF FIRST.CODE and RESULT='A' THEN OUTPUT;
ELSE IF LAST.CODE and RESULT NE 'A' THEN OUTPUT;
RUN;
So we are grouping the data by ID and CODE, and then writing to the dataset if certain conditions are met. I want to write a hive query to replicate this. This is what I have:
proc sql;
create table temp as
select *, row_number() over (partition by ID, CODE) as rowNum
from table1;
create table temp2 as
select a.ID, a.CODE, a.RESULT, a.RESULT2, a.RESULT3
from temp a
inner join (select ID, CODE, max(rowNum) as maxRowNum
from temp
group by ID, CODE) b
on a.ID=b.ID and a.CODE=b.CODE
where (a.rowNum=1 and a.RESULT='A') or (a.rowNum=b.maxRowNum and a.RESULT NE 'A');
quit;
There are two issues I see with this.
1) The row that is first or last in each BY group is entirely dependant on the order of rows in table1 in SAS, we aren't ordering by anything. I don't think row order is preserved when translating to a hive query.
2) The SAS code is taking the first row in each BY GROUP or the last, not both. I think that my HIVE query is taking both, resulting in more rows than I want.
Any suggestions or insight on how to improve my query is appreciated. Is it even possible to replicate this SAS code in HIVE?
The SAS code has a by statement (BY ID CODE;), which tells SAS that the set dataset is sorted at those levels. So, not a random selection for first. and last..
That said, we can replicate this in HIVE by using the first_value and last_value window functions.
FIRST.CODE should replicate to
first_value(code) over (partition by Id order by code)fcode
Similarly, LAST.CODE would be
last_value(code) over (partition by Id order by code)lcode
Once you have the fcode and lcode columns, use case when statements for the result column criteria. Like,
case when (code=fcode and result='A') or (code=lcode and result<>'A')
then 1 else 0 end as op_flag
Then the fetch the table with where op_flag = 1
SAMPLE
select id, code, result from (
select *,
first_value(code) over (partition by id order by code)fcode,
last_value(code) over (partition by id order by code)lcode
from footab) f
where (code=fcode and result='A') or (code=lcode and result<>'A')
Regarding point 1) the BY group processing requires the input data to be sorted or indexed on BY variables, so though the code contains no ordering, the source data is processed in order. If the input data was not indexed/sorted, SAS will throw error.
Regarding this, possible differences are on rows with same values of BY variables, especially if the RESULT is different.
In SAS, I would pre-sort data by ID, CODE, RESULT, then use BY ID CODE in order to not be influenced by order of rows.
Regarding 2) FIRST and LAST can be both true in SAS. Since your condition for first and last on RESULT is different, I guess this is not a source of differences.
I guess you could add another field as
row_number() over (partition by ID, CODE desc) as rowNumDesc
to detect last row with rowNumDesc = 1 (so that you skip the join).
EDIT:
I think the two programs above both include random selection of rows for groups with same values of ID and CODE variables, especially with same values of RESULT. But you should get same number of rows from both. If not, just debug it.
However the random aspect in SAS code/storage is based on physical order of rows, while the ROW_NUMBERs randomness within a group will be influenced by the implementation of the function in the engine.

Oracle pagination ROWNUM column>=value challenge

Having some trouble with oracle pagination. Case:
Table with > 1 billion rows:
Measurement(Id Number, Classification VARCHAR, Value NUMBER)
Index:
ON Measurement(Value)
I need a query that gets the first match and the following 2000 matches ordered by Value. I also would like to use the index.
First idea:
SELECT * FROM Measurement WHERE Value >= 1234567890
AND ROWNUM <= 2000 ORDER BY Value ASC
Result:
The query just returns the first 2000 cases it can find in the table, starting from the top, where Value is higher or equal to 1234567890, and then orders that resultset ascending.
Second idea:
SELECT * FROM
(SELECT * FROM Measurement WHERE Value >= 1234567890 ORDER BY Value ASC)
WHERE ROWNUM <= 2000
Result:
Oracle does not understand that ROWNUM should limit the amount from the inner query, so oracle decides to get all rows where Value is greater or equal to 1234567890 first, and then order that giant resultset before returning the first 2000 rows. Because Oracle is guessing that most of the data in the table will be returned, it ignores any use of index as well.
None of these approaches are acceptable as the first one gives the wrong results, and the second one takes hours.
Is pagination supported at all in Oracle?
You can use the following
SELECT * FROM
(SELECT Id, Classification, Value, ROWNUM Rank FROM Measurement WHERE Value >= 1234567890)
WHERE Rank <= 2000
order by Rank
You do not need to order in the sub-query. Simply unnecessary.
The above is not pagination but the firs page I would suppose.
Not sure if you got the solution for your problem, but to put my two cents:
The first query will not answer your requirements as it will fetch 2000 random records that satisfy your query and then do an order by.
Coming to the second query :
Oracle will first do the execution of the second query and will then only move to the outer query. So, the rownum filter will be applied only after the inner query is executed.
You can try the below approach, to do INDEX FAST FULL SCAN, i have tested it on a table with 2.76 million rows and it is having lesser cost than the other approach:
SELECT * from Measurement
where value in ( SELECT VALUE FROM
(SELECT Value FROM Measurement
WHERE Value >= 1234567890 ORDER BY Value ASC)
WHERE ROWNUM <= 2000)
Hope it Helps
Vishad
I think I have fond a potential solution. However, it's not a query.
declare
cursor c is
SELECT * FROM Measurement WHERE Value >= 1234567890 ORDER BY Value ASC;
l_rec c%rowtype;
begin
open c;
for i in 1 .. 2000
loop
fetch c into l_rec;
exit when c%notfound;
end loop;
close c;
end;
/
Kindly experiment with more options
SELECT *
FROM( SELECT /*+ FIRST_ROWS(2000) */
Id,
Classification,
Value,
ROW_NUMBER() OVER (ORDER BY Value) AS rn
FROM Measurement
where Value > 1234567889
)
WHERE rn <=2000;
Update1:- Force the use of index on Value.Here IDX_ON_VALUE is the Name of the index on Value in Measurement
SELECT * FROM
(SELECT /*+ INDEX(a IDX_ON_VALUE) */* FROM Measurement
a WHERE value >=1234567890 )
ORDER BY a.Value ASC)
WHERE ROWNUM <= 2000

Using rownum in subquery

In an algorithm the users passes a query, for instance:
SELECT o_orderdate, o_orderpriority FROM h_orders WHERE rownum <= 5
The query returns the following:
1996-01-02 5-LOW
1996-12-01 1-URGENT
1993-10-14 5-LOW
1995-10-11 5-LOW
1994-07-30 5-LOW
The algorithm needs the count for the select attributes (o_orderdate, o_orderpriority in the above example) and therefore it rewrites the query to:
SELECT o_orderdate, count(o_orderdate) FROM
(SELECT o_orderdate, o_orderpriority FROM h_orders WHERE rownum <= 5)
GROUP BY o_orderdate
This query returns the following:
1992-01-01 5
However the intended result is:
1996-12-01 1
1995-10-11 1
1994-07-30 1
1996-01-02 1
1993-10-14 1
Any idea how I could rewrite the parsing stage or how the user could pass a syntactically different query to receive the above results?
The rows returned by the inner query are essentially non-deterministic, as they depend on the order in which the optimiser identifies rows as part of the required data set. A change in execution plan due to modified predicates might change the order in which the rows come back, and new rows added to the table can also change which rows are included.
If you always want n rows then either use distinct(o_orderdate) in the innerquery, which will render the GROUP BY useless.
Or you can add another outer select with rownum to get n of the grouped rows, like this:
select o_orderdate, counter from
(
SELECT o_orderdate, count(o_orderdate) as counter FROM
(SELECT o_orderdate, o_orderpriority FROM h_orders)
GROUP BY o_orderdate
)
WHERE rownum <= 5
Although the results will most likely be useless as they will be undeterministic (as mentioned by David Aldridge).
As your outer query makes no use of "o_orderpriority", why not just get rid of the subquery and simply query like this:
SELECT o_orderdate, count(o_orderdate) AS order_count
FROM h_orders
WHERE rownum <= 5
GROUP BY o_orderdate

Resources