Just a bullet list of things that separate Cognos Transformer from being a capable OLAP-tool in modern market. Neither of these problems will be solved, I think, and this will lead to Transfomer’s death in near future (about 5 years, imho, given huge userbase). That’s a pity, since it’s a nice tool, absolutely brilliant 10 years ago. But as data sizes grow, Transformer just cannot scale. Hyperion did a nice job of inventing ASO option when it faced volume challenges, Cognos decided all was good enough.

List:

1. No dimension attributes There’s no way to add attributes to dimension elements, only to model them as separate dimensions (or alternate hierarchies), which is very inefficient both for modeling and for reporting. For example, try to display all sales in stores with 200-300 m^2 square footage
2. No multiple languages support No way to keep a single cube for German and English users, only separate cubes for each language and therefore more processing overhead and administration burden
3. Very slow zero suppression See this post
4. limited hardware resource optimization Transfomer is essentially a single-threaded application and cannot use more than 2 cores for cube build. Even with top level CPUs it’s just not enough. You can build multiple cubes in parallel to optimize resource usage, but that doesn’t help much either (see following point)
5. No multiple-cubes partitioning Cube processing limitations could be solved by dividing cube in multiple ‘cuboids’ and presenting them a single cube to end-user applications. However, time-partitioning imposes a serious limitation of only one time-dimension in cube and group paritioning doesn’t present a single cube model for analysis
6. 32bit application Transformer is a 32bit application and cannot utilize a lot of memory for processing, which could help things a lot. Moreover, famous cube file size limitation comes from this reason as well
7. Not MDX-complaint You can’t use anything except Cognos BI to work with Transformer cubes data. An XML\A provider would be very usefull
8. No Parent-Child hierarchies You can model then by dimension views but it requires some serious scripting. A very nice thing to have ‘out-of-the-box’

Given all of the above, powerplay is a nice single-user tool with friendly modeling gui or an OLAP-server in ‘small-data’ enviroment. As a single user tool it will be replaced by PowerPivot and as an OLAP-server it will be replaced by TM1 (which I’m not happy with — see following posts) or Microsoft Analysis Services.