(1) Explain your problem, don't simply post "This isn't working". What were you doing when you faced the problem? What have you tried to resolve - did you look for a solution using "Search" ? Has it happened just once or several times?
(2) It's also good to get feedback when a solution is found, return to the original post to explain how it was resolved so that more people can also use the results.
We work with a project containing about 40,000 activities at any one time. We like to Schedule ... (F9/reflow) the network regularly as logic ties and constraints are changed but the wait time is 30 to 40 seconds. Our IT folks tells us nothing can be done to speed it up.
Just curious if any other users operate in a similar environment and what sort of response times they get.
Much thanks,
Kevin
Printer-friendly version
Send to friend
Kevin
I am glad to here that they are running this script regulary to clean things up. I question the statement that p6 is using 32 bit addressing even on a 64 bit machine.
He is correct when he says that "Whenever we open or copy large projects or delete a large number of activities from a project it consumes a lot of resource on the workstation" Just like windows based environment you need space to delete and copy information to and from.
Also what version of P6 are you running. I can tell you that version 15 is a lot FASTER than 6.2 or 8.2 I mean a lot.
Zoltan,
Apparently our IT guys were aware of this technique and perform it regularly. They also offered the following reply to my concerns on performance ...
We have checked with the dba and there have been no memory or cpu alerts for P6 in the recent past, so that would tend to rule out a database performance issue. We have had some eSOMS batch job issues, spawning multiple instances of jobs, but none of those events has caused a resource shortage in the P6 instance.
More likely the thick-client is the culprit. Whenever we open or copy large projects or delete a large number of activities from a project it consumes a lot of resource on the workstation. This is an inherent limitation of the “thick-client” technology. P6 is certified to be able to handle 100,000 activities but as anyone will tell you that number is absolutely impractical due to how slow the network will load and perform. As the daily project approaches 40,000 activities we would expect that we would experience slow-downs in some of our basic operations. We have tested smaller projects in the production environment and response seems appropriate. Opening the daily project takes us approximately 40 seconds and that has been the norm for many years. P6 uses 32-bit addressing, even on a 64-bit machine so there isn’t much we can do from a technology standpoint to boost speed on data intensive operations.
have IT run that script that I posted. You can also try to press F10 or file commit changes then run the calculations.
If resource leveling is done manually I wonder how you manage your resource leveling. Manual leveling can be done by creating temporary soft links between activities competing for the same resources, another option can be moving the activities with the use of date constraints or perhaps you have another approach.
Maybe if having a lot of date constraints this adds overhead to the calculations of the network as the software will look for logic driven early dates even if to find most of them equal to Data Date in the absence of predecessor links and then will compare against date constraints to determine if constraints will be driving the start of the activities.
I do not like the idea of manually resource leveling a 40,000 activities network, but this is your option, so I am interested into finding out if there is a way to improve the calculation time for your benefit.
Good Luck
Rafael,
Sadly, resource leveling is done manually, so that's not a a contributor to processing time.
Resources are assigned at the group level with about 20 different types in use. Many of the activities are stand-alone with no logic so we typically run in the neighborhood of 20,000 to 30,000 logic ties at any one time.
Resource leveling a 40,000 activities schedule by hand must be a monumental task so I suppose the schedule run includes resource leveling.
I would like to know a bit more about the schedule like the number of logic links and the number of defined resources [perhaps in the hundreds or thousands].
Thanks Z. I don't have access to the backend, which I believe is Oracle, but I'll pass this on to our admin.
if you are using sql as you underlaying database run this script
select new query
use pmdb$primavera
update statistics project with fullscan
update statistics task with fullscan
then right click and select execute
In the "Messages" Window you should see "Command completed successfully". then run this scriptexec sp_updatestats then right click and select execute then exit
if you are using sql as you underlaying database run this script
select new query
use pmdb$primavera
update statistics project with fullscan
update statistics task with fullscan
then right click and select execute
In the "Messages" Window you should see "Command completed successfully". then run this scriptexec sp_updatestats then right click and select execute then exit
Kevin,
I think that 30-40 seconds for calculating 40000 activities schedule is not too bad, MSP is much slower.
Email: PPadmin@planningplanet.com
Replies