First of all this is a technical site (I like very much) and not a linguistical one. The really aim of that is to spread all the infos around.
Look. I'm Hungarian, I have never learned English or other languages and today I speek and write (more or less) English, German, and French. And I wouldn't be happy to see here on this site somebody criticize my English. This is just not the right place for that.
I think we would need to see much more articles like this, just to learn from them.
My personel opinion of clustering is following:
It is very, very hardware depending.
We have a qite big SAP infrastructure (100 GB) clustered on a kind of activ/activ cluster (the DB is on one node, the Central Instance on the other).
From the beginning on we have had a lot of problems with the cluster: Failovers, then after a fail over no fail back etc.
At the end our customer asked us to swich off (!) one node of the cluster during a critical year closing period because he has felt himself much more comfortable when the SAP was not clustered.
In the meantime the vendor has investigated the whole HW environment, exchanged a lot of HW pieces, but has never admitted the rootcause of the problems was the HW.
Now I have disk subsystem controller, cash, FC switch GBIC changed and now by wonder everything is working fine. So the vendor has closed the case.
So, to summerise it: cluster is fine if the HW, the OS, the patches, the BIOS for every pieces are up to date and cluster ready.