Our MAS Single plan is a single server deployment designed for up to 50* concurrent users. Our MAS Cluster plan includes a minimum of 2 worker nodes so can handle more users and, while not a “high availability” architecture distributed across multiple regions of the country, provides a level of resilience since it has more than one application server.
To estimate your Maximo concurrent users, find your current high-water mark with the following SQL query. This will return the top 10 concurrent user counts and the day each occurred in the prior year. You may choose to disregard any outliers.
You may also want to decrease your high-water mark if you have set Maximo’s time-out period to be longer than the standard 30 minutes, as this means the high-water number returned by the query may be inflated in terms of actual load on the server.
SQL Server:
select top 10 cast(attemptdate as Date) as countdate, max(currentcount) as concurrentcount from logintracking where attemptdate > getdate()-365 group by cast(attemptdate as Date)
order by max(currentcount) desc;
Oracle:
select * from
(select trunc(attemptdate) as countdate, max(currentcount) as concurrentcount from logintracking where attemptdate > sysdate-365 group by trunc(attemptdate)
order by max(currentcount) desc)
Where rownum <=10;
* IBM and Red Hat suggest that our MAS Single plan is appropriate for up to 70 concurrent users. We use 50 as the recommended number to be safe.