Richard L. Hamilton wrote:
> their own proprietary remote filesystem, not NFS).  As such, I wouldn't
> see the need to have an over-the-network block device like iSCSI for
> diskless swapping on Solaris.
>
>   
Fair enough - I thought it might be faster.
> I'd prefer to use the established way of doing it rather than fool with iSCSI.
> Nor can I see why one would even bother with (non-bootable software-based)
> iSCSI (rather than NFS) at all for a diskless Solaris system, unless one was
> running a DBMS that wanted a block device on them, in which case I'd
>   
My expectation is that for a primarily readonly filesystem, using the 
local buffer
cache against a block device should be quite a lot more efficient that 
using NFS.

> Since SPARC can boot over NFS without any local storage at all (and x86 could
> probably do it with reasonably small flash storage for grub and a miniroot),
> I can't see going that route very often.  The only use I see for iSCSI 
> initiator
> on Solaris is if one already has an iSCSI infrastructure (or has compelling
> reason to contemplate that rather than an NFS server).   (OTOH, using
>   
After the best part of 2 decades suffering NFS performance for software 
development
I'm a bit sick of it and keen to see if something more performant (where 
the fs is readonly,
or private to the workstation) is available.  I can see the point of NFS 
for shared areas
and home areas.  With a SVN server and an IMAP server, its not clear how 
much shared
space is really needed - some, but not much.
> So I'd say, in the case of Solaris, swap over NFS and just forget about iSCSI
> for swapping, even if it's _possible_, because absent an iSCSI HBA, you won't
> be _booting_ over iSCSI anyway, so you'll still need an NFS server.
>
>   
It looks like I might need one, yes.  I was expecting to use a PXE boot 
and then mount
the majority of the system from the server.  I know that NFS has made 
some some
advances in caching but my experiences (admitedly in environments where 
I don't control
the network environment) doesn't inspire confidence - and I *know* that 
the performance
against block devices is great.

Next problem is to work out how to strip down Solaris to see how small a 
system can
be and what's absolutely needed.
> Which approach is better for other operating systems, that's your problem...
>   
Well, maybe NFS and Solaris is the way to do it.  At the moment it looks 
like
diskless Linux and FreeBSD both have issues.  I'm surprised Sun don't make
more of this.  How much sense does it make to put disks on blades?  Or 
trader
desktops?  It makes deloyment more costly, uses more power, more to go 
wrong.
But we all buy blades and 1U servers with a pair of disks in them.  I'd 
rather
have another NIC (not a problem on some systems), more airflow, and a 
smaller
PSU.

My own interest is initially for diskless workstations (or at least X 
servers), though
I suspect diskless blades and compute farms are a bigger market, at 
least at the moment.

James

_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to