Re: Scheduled Transfer Protocol on Linux

From: Larry McVoy (lm@bitmover.com)
Date: Sat Feb 12 2000 - 18:59:27 EST


: > : over a bunch of disks it would be feasible, but if you are talking about
: > : single drives, this is sheer madness.
: >
: > To you, perhaps. I'll tell you this: I run a software business on Linux.
: > I can get 20GB drives for $200. If I could get 20GB drives with Linux
: > running on them for $300, I'd be buying them like cupcakes.
:
: Why? So you can log in to your disk drives and run emacs?

No, so I can log in and run vi :-)

The real reason is that I get network attached storage and I can have
any sort of service I want running on the drive. Perhaps you don't
see the whole picture. For my $300, I get a computer, not a disk
drive. I'm not plugging these drives into a machine, I'm plugging
them into the network. They are little servers.

Before you laugh at that, consider Cobalt's business. They sell little
slow machines that take very little space, and the fact that they are
slow just isn't an issue. They are fast enough and they solve some
problems people want solved.

: If there was some purpose that running Linux on a drive served, it might
: justify and extra $100 cost. I can't see any purpose to it. Yes, you could
: telnet into the thing and view your drive geometry, bad sector list,
: statistics, etc. You don't need to be running a GP/OS on your disk to get
: that information.

Huh? It's a COMPUTER. With a power cord and an ethernet cable. Run HTTP,
NFS, SMB, DNS, firewalling, whatever.

: I would much rather see that $100 going towards more storage capacity and
: buffering, rather than a CPU running linux. I think a lot of people would
: agree with me on that one.

Why do you think that?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Feb 15 2000 - 21:00:23 EST