|
|
Archives of the TeradataForum
Message Posted: Fri, 17 Jan 2003 @ 16:16:18 GMT
Subj: | | Re: Using Escon to move data |
|
From: | | Ray, Edmond |
OK I will lift the covers a little to explain my comments. The basic configuration of NPARC is that Teradata comes in and connects each
of the nodes on the source system to as many nodes as possible on the target system with direct connect Ethernet cables. In many cases this
requires adding network adapters. Creating as many individual networks between the system as possible. Then they have custom scripts to
configure, start, and monitor multiple TD load jobs across those networks. You probably would never want to run this on an operational
system, because normal operations would be delayed. Setting this up and making sure it works correctly is not a simple task.
You probably could do something similar with a single network between the systems using named pipes, and ARCMAIN, but it is not going to
be nearly as fast unless you are only running a single node system. And you still have to write and test the scripts.
It sounds like cost constraints may be limiting your choices, so best of luck, hope we can assist you in the future.
Edmond Ray
Senior Data Warehouse Consultant
Special Programs, NCR Government Systems
| |