wählt Durchschnitt der Top-3 für jeden spezifischen SQL-Auftrag

Question

SELECT jobs.Job_ID AS JobID , 
 
        jobs.NAME AS JobName , 
 
        act.start_execution_date AS StartExecutionDate , 
 
        AVG(FLOOR(run_duration/100)) AS AvgDurationMin , 
 
        CASE 
 
\t \t --If job average less than 5 minutes then limit is avg+10 minutes 
 
         WHEN AVG(FLOOR(run_duration/100)) <= 5 
 
         THEN (AVG(FLOOR(run_duration/100))) + 2 
 
\t \t --If job average greater than 5 minutes then limit is avg*limit percentage 
 
         ELSE (AVG(FLOOR(run_duration/100)) 
 
           ) 
 
        END AS DurationLimit , 
 
        DATEDIFF(MI, act.start_execution_date, GETDATE()) AS [CurrentDuration] 
 
      FROM @currently_running_jobs crj 
 
        INNER JOIN msdb..sysjobs AS jobs ON crj.job_id = jobs.job_id 
 
        INNER JOIN msdb..sysjobactivity AS act ON act.job_id = crj.job_id 
 
                   AND act.stop_execution_date IS NULL 
 
                   AND act.start_execution_date IS NOT NULL 
 
        INNER JOIN msdb..sysjobhistory AS hist ON hist.job_id = crj.job_id 
 
                   AND hist.step_id = 0 
 
      WHERE crj.job_state = 1 --and jobs.NAME = 'long_running_testing' 
 
      GROUP BY jobs.job_ID , 
 
        jobs.NAME , 
 
        act.start_execution_date , 
 
        DATEDIFF(MI, act.start_execution_date, GETDATE()) 
 
\t \t \t \t \t HAVING CASE WHEN AVG(FLOOR(run_duration/100)) <= 5 
 
\t \t \t \t \t \t \t THEN (AVG(FLOOR(run_duration/100))) + 2 
 
          --THEN (AVG(FLOOR(run_duration/100))) + 1 
 
         ELSE (AVG(FLOOR(run_duration/100)) 
 
           ) 
 
        END < DATEDIFF(MI, act.start_execution_date, GETDATE())

Kann jemand bitte helfen Sie mir die durchschnittliche run_duration der letzten drei Einträge der Aufgabe, durch job_name partitioniert zu extrahieren?

Das folgende Skript, das ich verwendet habe, um Informationen der letzten drei run_duration partitioniert von job_name, aber es berechnet den Durchschnitt aller run_duration.

Hier brauche ich nur die letzten drei Run_Duration Durchschnittszahl.

SELECT top 1 with ties jobs.Job_ID AS JobID , AVG(FLOOR(run_duration/100%100)), 
        jobs.NAME AS JobName , 
        act.start_execution_date AS StartExecutionDate 
from msdb..sysjobs AS jobs 
inner join msdb..sysjobactivity AS act ON act.job_id = jobs.job_id 
INNER JOIN msdb..sysjobhistory AS hist ON hist.job_id = jobs.job_id AND hist.step_id = 0 

GROUP BY jobs.job_ID , 
        jobs.NAME 
      ,  act.start_execution_date ,hist.run_duration, 
        DATEDIFF(MI, act.start_execution_date, GETDATE()) 
order by case when row_number() over (partition by jobs.NAME order by hist.run_duration desc) <= 3 then 0 else 1 end; 

Answer 1

Vielleicht ein CTE würde es vereinfachen.

with SeqByJob_cte as ( SELECT jobs.NAME AS JobName 
          ,hist.run_duration 
          ,row_number() over(partition by jobs.NAME order by act.start_execution_date desc) as seq 
        from msdb..sysjobs AS jobs 
          inner join msdb..sysjobactivity AS act ON act.job_id = jobs.job_id 
          INNER JOIN msdb..sysjobhistory AS hist ON hist.job_id = jobs.job_id AND hist.step_id = 0) 
select avg(run_duration) as AvgOf3MostRecentRuns 
     ,CASE 
    --If job average less than 5 minutes then limit is avg+10 minutes 
        WHEN AVG(FLOOR(run_duration/100)) <= 5 
        THEN (AVG(FLOOR(run_duration/100))) + 2 
    --If job average greater than 5 minutes then limit is avg*limit percentage 
        ELSE (AVG(FLOOR(run_duration/100)) 
          ) 
       END AS DurationLimit 
     ,JobName 
from SeqByJob_cte 
where seq <= 3 
group by JobName