9fans - fans of the OS Plan 9 from Bell Labs
 help / color / mirror / Atom feed
From: erik quanstrom <quanstro@quanstro.net>
To: 9fans@9fans.net
Subject: Re: [9fans] information about filesystems
Date: Fri, 13 Aug 2010 13:36:05 -0400	[thread overview]
Message-ID: <fb830975336cf155c895dde4eb15ebda@plug.quanstro.net> (raw)
In-Reply-To: <AANLkTi=v7UfhPPyYAzWkdCF17n2n5zoZ1iqtdBYi0xrG@mail.gmail.com>

> is there any source where I could read about differences between kfs
> and fossil (also I heard about something like cwfs), about reasons why
> fossil was designed and superceded, when to use which and alike?

steve's summary is excellent.  a few more words about my favorite,
ken's fs

the "worm" partition doesn't really need to be a worm.  it can
be any sort of device.  i currently use aoe targets.  typically coraid
appliances. but i have a few old fses using scsi disks or ata disks
hanging off ide/marvell/ahci controllers.

the big advantage of ken's fs is speed.  at coraid we top out at
>300mb/s file io on a lowly intel 5000 xeon and a old coraid
appliance.  ken's fs is able to serve so quickly because
- it doesn't waste time compressing (compression is slower than
disks these days!)
- it can have a huge memory cache — nearly all of memory.
- it doesn't need to copy between user space and kernel space;
it also doesn't do anything with the mmu except set up 1024
4mb pages on startup.
- it keeps files less fragmented, due to fact that it doesn't
ensure that blocks are globally unique.

ken's fs also tends to be stable enough to be externally
uptime limited.  things like ups batteries and network hiccups
tend to be more disruptive.

here are a few more ken's fs references.
	http://www.quanstro.net/plan9/disklessfs.pdf
	http://www.quanstro.net/plan9/history.pdf

- erik



      parent reply	other threads:[~2010-08-13 17:36 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-13 16:11 Rudolf Sykora
2010-08-13 16:31 ` Steve Simon
2010-08-13 17:26   ` Rudolf Sykora
2010-08-26 19:42   ` Ethan Grammatikidis
2010-08-26 23:53     ` erik quanstrom
2010-08-13 17:36 ` erik quanstrom [this message]

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=fb830975336cf155c895dde4eb15ebda@plug.quanstro.net \
    --to=quanstro@quanstro.net \
    --cc=9fans@9fans.net \
    /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).