|
|
Archives of the TeradataForum
Message Posted: Thu, 13 Apr 2006 @ 18:53:39 GMT
Subj: | | Re: Experiences with TSM and Teradata |
|
From: | | Meade, Cornelius |
In a prior life I used to evaluate new vendor hardware for the company I then worked for and one of my areas of responsibility was tape drive
hardware as we were always in the market back then for the fastest/best tape drives hardware we could get our hands on. This was many years ago
(more than I'd care to say!) but the point of this story is to convey that during that time it was fairly common to observe an effect similar to
the one the original poster described. I had so called "saturation" test procedures of various types I would run whose main purpose was to
achieve the maximum data transfer though all of the subordinate drives. The problem effect we often found with these tests was generally due to
throughput limitations in the device controller that was managing a set of tape drives. There were various ways or root causes (translation:
shortfalls in the construction and/or microcode of the controller section) for the problem but the outward effect in almost all cases was
identical to the one described. Of course, there could be lots of other factors at play here but the effect the poster observed would certainly
pique my curiosity also.
Cornelius Meade III
EDS - Consulting Services/BI/CRM
| |