Also Sprach Joshua Baker-LePain:
> On Fri, 6 Jul 2001 at 10:37pm, C. Chan wrote
>
> > In short there is no simple answer.
> >
> Exactly. Let me add one more thing to consider, and that is vendor
> reliability.
This might be better described as vendor viability - how likely
is it a vendor will exist 3-5 years down the line? Past that point
you'll likely end considering migrating to new technologies, just
as DDS and 8x00 8mm gave way to DLT, and DLT is being supplanted
by SuperDLT, LTO, AIT-N.
Vendor reliability is how likely they can repair and return
a drive to you on time, or to resolve firmware and other
compatibility problems.
> Some groups here were recently testing AIT2 vs. Mammoth2.
> They liked Mammoth2, but went with AIT2 because a) it offered similar
> performance, and b) from the rumblings they heard, Exabyte as a company
> may be on the ropes. Exabyte took a long time to bring Mammoth to market,
> and don't have much of a roadmap for the technology (or for themselves).
> Sony, on the other hand, isn't going away any time soon, and has a solid
> roadmap for AIT, extending, IIRC, through AIT6.
>
Judging by vendor size and success in the marketplace, I expect
both SuperDLT and LTO to be around in 5 years; which will dominate
will be a matter of marketing and compatibility. AIT also has a
good future as the premier 8mm technology.
Among smaller companies, one gauge of success will be the OEM
contracts. The recent successes of Ecrix and Benchmark, and the failure of
OnStream were due to OEM contracts. In retrospect Ecrix has survived because
its price point makes it very attractive to those migrating
from DDS, Travan, etc. and the 8mm form factor makes retooling
for autoloaders and jukeboxes unnecessary. The OEM contract with
Compaq is a definite boost. The same comments hold for DLT1.
OnStream on the other hand had the attractive price point but
its form factor as well as some reliability and compatibility problems
prevented its entry into the OEM space. I expect DLT1 to survive
as an entry point linear tape. I don't know if Ecrix will survive
as a separate company but its packet write technology may find its
way into the next generation of tape.
Tandberg SLR is a mature proven technology but Tandberg now sells DLT as
well and SLR may get phased out in favor of DLT. I expect IBM Magstor to
persist since it fills a niche no other linear tape fills.
I also fully to expect to see tape RAID or RAIT, filesystem spanning
across tapes, multiplexing of I/O streams a la Legato and Syncsoft
in Amanda over the next 5 years :-).
--
C. Chan < [EMAIL PROTECTED] >
Finger [EMAIL PROTECTED] for PGP public key.
