Archives of the TeradataForum
Message Posted: Sat, 14 Jun 2003 @ 20:46:29 GMT
What you're trying to prove is not a whole true. SPARC didn't survive just because of lack of M$ alternatives - this architecture, with all improvements comparing to traditional RISC (like special registers to manage DMA), was just better than Alpha or MIPS at this time.
Also, I'm sure there is still place for solutions, based on own IP chips - Teradata/NCR solutions are great, but based on a very old solutions. If others (I don't claims it have to be Netezza) are able to build MPP, vector, multiflow or ccNUMA machine, able to do particular task 10 times faster, I'm asking - why not? And why such a vendor have to wait "some years" to prove it's value?
|Copyright 2016 - All Rights Reserved|
|Last Modified: 27 Dec 2016|