Hi Folks,
What is the affect of 1 million cells per elist item?
And how do we calculate the cells per elist item?
I would appreciate if someone explains with an example... :)
And also how we increase the performance when model size is beyond 1 million cells?
Thanks
Preethi
:)
Hi Peethi,
You can see the exact number of cells per elist item in Contributor Administration Console under Development>Application Maintenance>Application Details.
Although 1 million cells is usually still workable/acceptable for end users in terms of performance, that is for me personally the passmark where I start thinking about options to improve performance.
There are many ways to improve performance and most of them are depending on the model, the design and the requirements.
One is the previous discussed cut down option, but that should be the last resort and only be used if it can't be solved with a different design.
Design changes can be for example a different dimension as elist or the incorporation of another DList in the elist. But that depends on the required user access, planning workflow process, etc.
Check which cubes have a big sparcity (lot of empty cells) and focus on these.
Another option is to use DList formatted items as "virtual dimensions". For example if you have 1200 employees in a company, but a maximum of 50 employees per cost center (eList). You can use a DList with number 1 to 50 and include the employee name or number as a text field in another dimension.
Or combining DList if you have a unique relationship (hierarchies). For example Products and Product Groups . Assuming that each Product would belong only to one Product Group, you would use PG as subtotals in the Product hierarchy. Obvious, but you get the concept and it could be applied to other scenarios, i.e Customers and Regions (if unique, etc.)
Hope that helps.
Carsten
I was told the limit for each e-list is 1.4Million. Theorectically 1 Million cell is technically feasible
Cheers
Vincent
I am just curious - why do you think the limit is 1.4M ?