too my understanding, postgres is now architected so that engine and storage are ""splited"" and so one may think that storage code different that the traditional postgres thing is possible. I have heard of oracle like redo log things, ..etc
I wonder if a memory thing is on the way, somewhere ? The why (1) of this question: its quite easy, nowadays to find blades with 2 TB of RAM. (or even much more !) and, to my understanding (correct me if I am wrong !) postgres shared buffers are not really efficient with "some" Terabyte of allocated space. The why (2) I may have to do a POC with Sap HANA on one side and postgres on the other on machines with a lot of RAM. so... can someone shade some light ? thanks, Marc MILLAS Senior Architect +33607850334 www.mokadb.com
