Experience with DataMining in 3.5 ?

[Former Member]

We have a datamining project. I'm wondering who has experience with datamining in BW3.5 (except ABC classification which is really basic). What kind of issues we can have. What models (clustering...) are really user.

For example performance are very bad with small volume (200lines). I'm wondering what the performance will be in prod.

Many thanks

Philippe

[former_member93896 · Accepted solution]

Hello Philippe,

I will let others comment on their experience, just keep in mind that this is relatively new compared to the rest of BW.

200 records should not be a problem at all when it comes to performance. In which step/function is the system slow? If it is the first step (data selection), then it is most likely not the fault of data mining. You should check database parameters and statistics.

Regards,

Marc

SAP NetWeaver RIG, US BI