(Moving this to discuss, since it no longer has anything to do with actual
coding whatsoever. ;-)
>
> But, true, the lack of enthusiasm and help (and money) from
> these boxes owners may be a showstopper, indeed.
*That* is the clincher. There are tons and tons of SPARC boxes still around.
But *nobody* has stepped forward with a plan to invest in what everyone
recognizes is ultimately a dead end. The current box owners are basically the
folks (e.g. Unis) that got this stuff for free. Generally they reason they
are still using this gear is precisely because they lack either the money to
buy equipment, or the skills to port, or both. Which generally makes their
ability to be anything other than freeloaders somewhat questionable. So the
challenge is to figure out how supporting those guys servers a bigger, more
realistic, business opportunity. Nobody has presented me with a reasonable
business case yet, although a couple of people have *tried*.
Right now the trend for folks with money is to move off that gear, and mostly
into cloud-like deployments.
There is a non-zero SPARC opportunity still.
But's pretty small, and rapidly diminishing. Even two years ago, I think
there may have been some value. But now -- I just don't see it. Who here has
*SPARC* hardware in their data center, is wanting a newly updated operating
system for it, and is willing to pay significantly for such. (To be
commercially interesting, I think there would need to be either huge numbers of
such people, or people willing to pay support contracts that are similar to
what Oracle charges. I can tell you that it certainly doesn't make sense for
me to work to support even 100 customers who don't want to spend more than $500
a machine for software support. I think to be a viable business, you'd need to
come up with at least $1M in commitments, with probably a substantial portion
of that *upfront*. That would let you hire a few guys, and get the equipment
you need to support them, including footing the power bill. If you can't get
at least that much, then the opportunity probably simply isn't worth it.)
And the "interesting" SPARC hardware (e.g. E10K) takes significant resources to
support -- we don't have all the source for it, for example. Mostly the stuff
we can support *right now* is desktop and lower end server (e.g. E450 class)
hardware. We don't have complete source to support anything called "SunBlade"
for example. (And there are no graphics options left for those either, except
some really really lame 2D cards. None of the 3D cards was ever open sourced.
And for that matter, supporting X properly on SPARC workstations is probably a
much larger task than anyone here would want to do -- with exactly zero
commercial benefit.)
- Garrett
-------------------------------------------
illumos-discuss
Archives: https://www.listbox.com/member/archive/182180/=now
RSS Feed: https://www.listbox.com/member/archive/rss/182180/21175430-2e6923be
Modify Your Subscription:
https://www.listbox.com/member/?member_id=21175430&id_secret=21175430-6a77cda4
Powered by Listbox: http://www.listbox.com
