Many thanks for your replies so far on this. The issue I am talking about is the same one mentioned by Paul at the LinkedIn URL above - and it is also the subject of a CA Support call. As Paul explains in that link, we are seeing an error when trying to add an additional attribute for inclusion in the DWH. The error seen in the log file tells us that the view ODF_PROJECT_V can't be rebuilt as it has exceeded the maximum number of columns (1024 for SQL server).
Once we have reached the limit in terms of DWH selected attributes, any run attempt of the job "Load Data Warehouse" gives us the error:-
Couldn't execute SQL: MERGE INTO DWH_INV_PROJECT TGT
[CA Clarity][SQLServer JDBC Driver][SQLServer]Invalid column name 'ZZ_IBS_BFA7'.
Looking at the contents of ODF_PROJECT_V, it would appear that the number of columns in the view rises in proportion to the number of attributes on the project object. For an OOTB 14.3 instance there are 230 project object attributes - and 483 columns in the table. The number of columns in the table then increases according to the type of each new custom attribute added - for example each Boolean attribute adds 5 view columns - others create just one or two. Any DWH-synced attribute that is added to the tables in INV_INVESTMENT, INV_PROJECT, FIN_FINANCIALS, PAC_MNT_PROJECTS, ODF_CA_PROJECT or ODF_CA_INV - seems to get added as columns on this view. If this assumption is correct then - as Suman points out - this would seem to indicate that there is only room to add in about 100 custom attributes on the project and/or investment objects before we hit problems. That is only about a 10th of the number of custom attributes which we could have prior to v 14 and could be a major architectural problem to anybody upgrading such a system.
It may be, as Nika points out, that the ODF_PROJECT_V view does not directly affect the Load Data Warehouse job - but it would seem that it could have secondary effects once a given number of synced project/investment attributes have been created.
So my question here is all about trying to disprove our theory by looking for anybody who was in a similar position to us - but who had not ran into problems. Still waiting for any replies on that score.