9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: "Francisco J Ballesteros" <nemo@lsub.org>
To: "Fans of the OS Plan 9 from Bell Labs" <9fans@cse.psu.edu>
Subject: Re: [9fans] QTCTL?
Date: Thu,  1 Nov 2007 19:01:41 +0100	[thread overview]
Message-ID: <8ccc8ba40711011101l550e021dgf7f5379abc616627@mail.gmail.com> (raw)
In-Reply-To: <8ccc8ba40711011056p2be8772dm4f061d5a401445ce@mail.gmail.com>

One thing that Op taught us was that it´s quite nice to add changes in
the protocol
just by NOT changing the protocol, and putting special servers-clients
as gateways.

We could implement a export for caching clients (and a caching client) so that
the export implements the Tinval-Rinval thing, and serves the caching clients.
The export process may export its own namespace (perhaps just with the main
fileserver mounted) and the caching clients would behave as 9p servers to others
in their machines. This worked just fine in Inferno and could work if
we try it in Plan 9.

Also, I´m now thinking that we could take advantage of these new
xport-cache links
to avoid some RPCs. To maintain 9p semantics we could not use op in there, but
perhaps there´s something in the middle. I have to give this all more thought.

Any comment on this, though?


On 11/1/07, Francisco J Ballesteros <nemo@lsub.org> wrote:
> We did look at nfs v3 and lbfs before implementing op.
>
> nfs v3 at least tried to address latency and not just bandwidth as lbfs,
> but it seemed to use more RPCs than needed for some tasks (I don´t remember
> now, but have that written down somewhere).
>
>
>
> On 11/1/07, ron minnich <rminnich@gmail.com> wrote:
> > right, but before the experiments start in earnest, see what they're
> > doing at lustre.org, in nfs v3, etc.
> >
> > Much of this discussion is familiar.
> >
> > <shameless plug> you can also see what I did in mnfs ca. 1992, if you
> > promise to ignore the use I put it to (DSM). I implemented invalidates
> > for shared pages. It took, like 15 minutes to implement. It required
> > that I run an nfs server on each client, however, and it worked
> > because NFS blocks on a node have a global name: <fhandle>:<offset>.
> > So the server tracked who had what pages, and an invalidate was
> > actually a simple RPC from servers to clients. yes, this broke the
> > c-s-c model, but hey ... I like it better than leases, personally.
> >
> > But neither this nor leases seems to scale terribly well to 4096 or
> > more clients.
> >
> > ron
> >
>

  reply	other threads:[~2007-11-01 18:01 UTC|newest]

Thread overview: 77+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-10-31 18:40 Francisco J Ballesteros
2007-10-31 18:56 ` Eric Van Hensbergen
2007-10-31 19:13   ` Charles Forsyth
2007-10-31 19:33     ` Eric Van Hensbergen
2007-10-31 19:39       ` erik quanstrom
2007-10-31 20:43     ` geoff
2007-10-31 21:32       ` Charles Forsyth
2007-10-31 22:48       ` roger peppe
2007-10-31 23:35         ` erik quanstrom
2007-11-01  9:29           ` roger peppe
2007-11-01 11:03             ` Eric Van Hensbergen
2007-11-01 11:19               ` Charles Forsyth
2007-11-01 12:11             ` erik quanstrom
2007-10-31 19:42 ` erik quanstrom
2007-10-31 19:49   ` Eric Van Hensbergen
2007-10-31 20:03     ` erik quanstrom
2007-10-31 20:10       ` Latchesar Ionkov
2007-10-31 20:12       ` Eric Van Hensbergen
2007-10-31 20:17       ` Russ Cox
2007-10-31 20:29         ` Francisco J Ballesteros
2007-10-31 20:48           ` Charles Forsyth
2007-10-31 21:23             ` Francisco J Ballesteros
2007-10-31 21:40               ` Russ Cox
2007-10-31 22:11                 ` Charles Forsyth
2007-10-31 22:26                 ` Francisco J Ballesteros
2007-10-31 22:37                   ` Charles Forsyth
2007-10-31 22:43                     ` Francisco J Ballesteros
2007-10-31 23:32                     ` Eric Van Hensbergen
2007-10-31 23:41                       ` [V9fs-developer] " Charles Forsyth
     [not found]                       ` <606b6f003ae9f0ed3e8c3c5f90ddc720@terzarima.net>
2007-11-01  1:13                         ` Eric Van Hensbergen
2007-10-31 23:54                   ` erik quanstrom
2007-11-01  0:03                     ` Charles Forsyth
2007-11-01  1:25                     ` Eric Van Hensbergen
2007-11-01  1:44                       ` erik quanstrom
2007-11-01  2:15                         ` Eric Van Hensbergen
2007-11-01  7:34                       ` Skip Tavakkolian
2007-11-01  6:21                 ` Bakul Shah
2007-11-01 14:28                   ` Russ Cox
2007-11-01 14:38                     ` erik quanstrom
2007-11-01 14:41                     ` Charles Forsyth
2007-11-01 15:26                     ` Sape Mullender
2007-11-01 15:51                       ` Latchesar Ionkov
2007-11-01 16:04                         ` ron minnich
2007-11-01 16:16                           ` Latchesar Ionkov
2007-11-01 16:21                           ` Sape Mullender
2007-11-01 16:58                             ` Francisco J Ballesteros
2007-11-01 17:11                               ` Charles Forsyth
2007-11-01 17:11                                 ` Francisco J Ballesteros
2007-11-01 17:13                               ` Sape Mullender
2007-11-01 17:38                               ` ron minnich
2007-11-01 17:56                                 ` Francisco J Ballesteros
2007-11-01 18:01                                   ` Francisco J Ballesteros [this message]
2007-11-01 18:52                                   ` Eric Van Hensbergen
2007-11-01 19:29                                     ` Francisco J Ballesteros
2007-11-01 23:24                                   ` ron minnich
2007-11-01 17:03                           ` Russ Cox
2007-11-01 17:12                             ` Sape Mullender
2007-11-01 17:35                               ` erik quanstrom
2007-11-01 18:36                                 ` erik quanstrom
2007-11-01 17:13                             ` Charles Forsyth
2007-11-01 17:16                               ` Charles Forsyth
2007-11-01 17:20                                 ` Charles Forsyth
2007-11-01 17:52                             ` Eric Van Hensbergen
2007-11-01 18:00                             ` Latchesar Ionkov
2007-11-01 18:03                               ` Francisco J Ballesteros
2007-11-01 18:08                                 ` Latchesar Ionkov
2007-11-01 18:16                                   ` erik quanstrom
2007-11-01 18:19                                   ` Francisco J Ballesteros
2007-11-01 18:35                                     ` Sape Mullender
2007-11-01 19:09                                       ` Charles Forsyth
2007-11-01 19:07                                         ` erik quanstrom
2007-11-01 17:14                           ` Bakul Shah
2007-11-01 16:17                         ` Sape Mullender
2007-11-01 16:27                           ` Sape Mullender
2007-11-01 16:58                         ` Sape Mullender
2007-11-01 16:59                     ` Bakul Shah
2007-11-01  4:21 Brian L. Stuart

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8ccc8ba40711011101l550e021dgf7f5379abc616627@mail.gmail.com \
    --to=nemo@lsub.org \
    --cc=9fans@cse.psu.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).