Mssql SQL query

Try..

SELECT * , COUNT(*) OVER (PARTITION BY YEAR([InsertedDate]),MONTH([InsertedDate]))  EventMonthCount
from ArchiveLoginsLogs
where [InsertedDate] >='2010-01-01' and [InsertedDate] <'2016-11-29' 

Select allOpen in new window

If you want sum of any column use below- SUM(YourColumnName)

SELECT * , SUM(YourColumnName) OVER (PARTITION BY YEAR([InsertedDate]),MONTH([InsertedDate]))  EventMonthCount
from ArchiveLoginsLogs
where [InsertedDate] >='2010-01-01' and [InsertedDate] <'2016-11-29' 

Select allOpen in new window

Hope it helps !!

for Vitor's comment, you no need to select [InsertedDate] ... and  group by [InsertedDate],

hence:

select year(InsertedDate) as 'Year', month(InsertedDate) as 'Month', count ([InsertedDate]) as EventCount
 from ArchiveLoginsLogs
 where [InsertedDate] >='2010-01-01' and [InsertedDate] <'2016-11-29' 
 group by year(InsertedDate), month(InsertedDate)  
 Order by 1,2

Select allOpen in new window

for Vitor's comment, you no need to select  [InsertedDate] ... and  group by [InsertedDate],

True. I just copied the code and forgot to update that part. Nice catch, Ryan.

Guys,
Both query did not  return me the data I want, the query below provide me with counts of data daily, but how do I sums this  EventCount so that I can present the count monthly ? so the output that I'm looking for shoud be like thsi .

InsertedDate      EventCount
2016-01-01      4216
2016-02-02      1925
2016-03-03      1854
2016-04-04      4606

Or can be like this

InsertedDate      EventCount
2016-01-01 - 2016-01-31      4216
2016-02-02  - 2016-02-29      1925
2016-03-03 - 2016-03-30      1854
2016-04-04 - 2016-03-31    4606




select [InsertedDate],count ([InsertedDate]) as EventCount
from ArchiveLoginsLogs
where [InsertedDate] >='2010-01-01' and [InsertedDate]
<'2016-11-29'
group by [InsertedDate]
Order by [InsertedDate] asc

short output from the query

InsertedDate      EventCount
2016-01-01      4216
2016-01-02      1925
2016-01-03      1854
2016-01-04      4606

Try..

SELECT DISTINCT DATEFROMPARTS(YEAR([InsertedDate]),MONTH([InsertedDate]),'01') InsertedDate , COUNT(*) OVER (PARTITION BY YEAR([InsertedDate]),MONTH([InsertedDate]))  EventCount
from ArchiveLoginsLogs
where [InsertedDate] >='2010-01-01' and [InsertedDate] <'2016-11-29' 

Select allOpen in new window

Or this...

SELECT DISTINCT CONCAT(  DATEFROMPARTS(YEAR([InsertedDate]),MONTH([InsertedDate]),'01') , ' - ' ,
EOMONTH(DATEFROMPARTS(YEAR([InsertedDate]),MONTH([InsertedDate]),'01')) )InsertedDate
,COUNT(*) OVER (PARTITION BY YEAR([InsertedDate]),MONTH([InsertedDate]))  EventCount
from ArchiveLoginsLogs
where [InsertedDate] >='2010-01-01' and [InsertedDate] <'2016-11-29' 

Select allOpen in new window

The periods shouldn't start all at day one? Why Feb starts at day two and March at day three and so on?
InsertedDate      EventCount
 2016-01-01 - 2016-01-31      4216
 2016-02-02  - 2016-02-29      1925
 2016-03-03 - 2016-03-30      1854
 2016-04-04 - 2016-03-31    4606

Also, if you want a SUM why the above example isn't a SUM but the values from January days?
InsertedDate      EventCount
 2016-01-01      4216
 2016-01-02      1925
 2016-01-03      1854
 2016-01-04      4606

SELECT DATEADD(MONTH, DATEDIFF(MONTH, 0, [InsertedDate]), 0) AS [InsertedMonth],
    COUNT(*) AS EventCount
FROM ArchiveLoginsLogs
WHERE [InsertedDate] >='2010-01-01' AND
              [InsertedDate] < DATEADD(DAY, DATEDIFF(DAY, 0, GETDATE()), 0)
GROUP BY DATEADD(MONTH, DATEDIFF(MONTH, 0, [InsertedDate]), 0)
ORDER BY DATEADD(MONTH, DATEDIFF(MONTH, 0, [InsertedDate]), 0)


Don't use string conversion in any way (unless absolutely unavoidable), because it will vastly slow down (and complicate) the query.

The general format above could be used for yearly totals also, for example:

SELECT DATEADD(YEAR, DATEDIFF(YEAR, 0, [InsertedDate]), 0) AS [InsertedYear],
...

Thus, this technique is not only very efficient but it's also consistent for different date granularities.  For example, if you had a full datetime and wanted hourly totals:
SELECT DATEADD(HOUR, DATEDIFF(HOUR, 0, [InsertedDate]), 0) AS [InsertedHour],
...

Thanks guys,I managed it with your guys help