TM1 Forum

Hi there ,

I am trying to optimize a data load process which takes 1 hour to load . I have used TM1RunTI and split the process according to a static dimension . Everything works well and loading the data is achieved in 12-13 mins . But the catch is before the data load resumes in Epilog tab the conditional feeders in the source cube is processed using CubeProcessFeeders in the Prolog Tab . So basically when i hash out the Cube Process feeders the data load happens in 12-13 mins but when i run the Cubeprocessfeeders function before the data load with TM1RunTI the same process takes 45 mins to complete .

I have tested and tried analyzing buy running the data load and cube process feeders individually :
CubeProcessfeeders takes 15 minutes
Data load takes 12-13 minutes.
But if they are run together in same process its taking 45 mins . Not sure why this lag.
Can someone give some insight on this ? It will be really helpful .

Thanks

Since you have the conditional feeders where the condition clause contains a fed value, may not be good with this approach. may not work in parallel interaction.

Why do you need (or think you need) CubeProcessFeeders?