All of lore.kernel.org
 help / color / mirror / Atom feed
From: Mike Hardy <mhardy@h3c.com>
To: linux-raid@vger.kernel.org
Subject: Re: swp - Re: ext3 journal on software raid (was Re: PROBLEM: Kernel 2.6.10 crashing repeatedly and hard)
Date: Wed, 05 Jan 2005 17:28:00 -0800	[thread overview]
Message-ID: <41DC9420.5030701@h3c.com> (raw)
In-Reply-To: <200501060057.j060vk928131@www.watkins-home.com>


Guy wrote:

> So, the question is:
> 	Assume I have enough RAM to never need to swap.
> 	Do I need any swap space with Linux?

This has been hashed out at great length on linux-kernel - with a few 
entrenched positions emerging, if I recall correctly.

There are those that think if they size the machine correctly, they 
shouldn't need swap, and they're right.

> I had 8 Meg of swap space.  So 12 meg total available virtual memory.
> One day I added 16 Meg of RAM.  So now I had 20 Meg of RAM.  I deleted my
> swap space.  Everyone told me I needed 20-40 Meg of swap space now!  Swap
> space should be 2 time RAM size.  How crazy, my memory requirements did not
> change, just the amount of memory.  I used that system for a year or so like
> that.  Go figure!

This story pretty much sums up that position, and I've certainly been in 
that position myself.

There are others (notably the maintainer of part of that code - Andrea 
Arcangeli if I recall correctly, though I apologize if that's a 
misattribution) who believe you should always have swap because it will 
allow your system to have higher throughput. If you process a large I/O, 
for instance, the kernel can swap out live processes to devote more RAM 
to VFS caching. That hurts latency though, and the nightly slocate run 
is a pathlogical example of this. You wake up in the morning and your 
machine is crawling while it swaps everything back in.

There's a "swappiness" knob you can twiddle in /proc or /sys to alter 
this, but the general idea is that you can improve throughput on a 
machine by having swap even if there is enough ram for all processes

I've got machines with and without swap, but I typically run all servers 
with swap to handle ram-usage spikes I'm not expecting (you never know) 
while I run my laptop without swap when possible to avoid latency issues 
with swappiness.

As with all things, its policy decisions and tradeoffs :-)

-Mike

  reply	other threads:[~2005-01-06  1:28 UTC|newest]

Thread overview: 104+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-12-30  0:31 PROBLEM: Kernel 2.6.10 crashing repeatedly and hard Georg C. F. Greve
2004-12-30 16:23 ` Georg C. F. Greve
2004-12-30 17:39   ` Peter T. Breuer
2004-12-30 17:53     ` Sandro Dentella
2004-12-30 18:31       ` Peter T. Breuer
2004-12-30 19:50     ` Michael Tokarev
2004-12-30 21:39       ` Peter T. Breuer
2005-01-02 19:42         ` ext3 journal on software raid (was Re: PROBLEM: Kernel 2.6.10 crashing repeatedly and hard) Andy Smith
2005-01-02 20:18           ` Peter T. Breuer
2005-01-03  0:30             ` Andy Smith
2005-01-03  6:41               ` Neil Brown
2005-01-03  8:37                 ` Peter T. Breuer
2005-01-03  8:03               ` Peter T. Breuer
2005-01-03  8:58                 ` Guy
2005-01-03 10:18                 ` Partiy error detection - was " Brad Campbell
2005-01-03 12:11                 ` Michael Tokarev
2005-01-03 14:23                   ` Peter T. Breuer
2005-01-03 18:30                     ` maarten
2005-01-03 21:36                     ` Michael Tokarev
2005-01-05  5:50                     ` Debian Sarge mdadm raid 10 assembling at boot problem Roger Ellison
2005-01-05 13:41                       ` Michael Tokarev
2005-01-05 13:57                         ` [help] [I2O] Adaptec 2400A on FC3 Angelo Piraino
2005-01-05 19:15                         ` Debian Sarge mdadm raid 10 assembling at boot problem Roger Ellison
2005-01-05  9:56           ` ext3 journal on software raid (was Re: PROBLEM: Kernel 2.6.10 crashing repeatedly and hard) Andy Smith
2005-01-05 10:44             ` Alvin Oga
2005-01-05 10:56               ` Brad Campbell
2005-01-05 11:39                 ` Alvin Oga
2005-01-05 12:02                   ` Brad Campbell
2005-01-05 13:23                     ` Alvin Oga
2005-01-05 13:33                       ` Brad Campbell
2005-01-05 14:44                         ` parts -- " Alvin Oga
2005-01-19  4:46                           ` Clemens Schwaighofer
2005-01-19  5:05                             ` Alvin Oga
2005-01-19  5:49                               ` Clemens Schwaighofer
2005-01-19  7:08                                 ` Alvin Oga
2005-01-05 13:36                       ` Swap should be mirrored or not? (was Re: ext3 journal on software raid) Andy Smith
2005-01-05 14:12                 ` ext3 journal on software raid (was Re: PROBLEM: Kernel 2.6.10 crashing repeatedly and hard) Erik Mouw
2005-01-05 14:37                   ` Michael Tokarev
2005-01-05 14:55                     ` errors " Alvin Oga
2005-01-05 17:11                     ` Erik Mouw
2005-01-06  5:41                       ` Brad Campbell
2005-01-05 15:17                 ` Guy
2005-01-05 15:33                   ` Alvin Oga
2005-01-05 16:22                     ` Michael Tokarev
2005-01-05 17:23                       ` Peter T. Breuer
2005-01-05 16:23                     ` Andy Smith
2005-01-05 16:30                       ` Andy Smith
2005-01-05 17:04                       ` swp - " Alvin Oga
2005-01-05 17:26                         ` Andy Smith
2005-01-05 18:32                           ` Alvin Oga
2005-01-05 22:35                             ` Andy Smith
2005-01-06  0:57                               ` Guy
2005-01-06  1:28                                 ` Mike Hardy [this message]
2005-01-06  3:32                                   ` Guy
2005-01-06  4:49                                     ` Mike Hardy
2005-01-09 21:07                                       ` Mark Hahn
2005-01-06  5:04                                   ` Alvin Oga
2005-01-06  6:18                                     ` Guy
2005-01-06  6:31                                       ` Alvin Oga
2005-01-06  9:38                                     ` swap on RAID (was Re: swp - Re: ext3 journal on software raid) Andy Smith
2005-01-06 17:46                                       ` Mike Hardy
2005-01-06 22:08                                         ` No swap can be dangerous (was Re: swap on RAID (was Re: swp - Re: ext3 journal on software raid)) Andrew Walrond
2005-01-06 22:34                                           ` Jesper Juhl
2005-01-06 22:57                                             ` Mike Hardy
2005-01-06 23:15                                               ` Guy
2005-01-06 23:15                                                 ` Guy
2005-01-07  9:28                                                 ` Andrew Walrond
2005-02-28 20:07                                                   ` Guy
2005-02-28 20:07                                                     ` Guy
2005-01-07  1:31                                       ` confused Re: swap on RAID (was Re: swp - Re: ext3 journal on software raid) Alvin Oga
2005-01-07  2:28                                         ` Andy Smith
2005-01-07 13:04                                           ` Alvin Oga
2005-01-09 21:21                                     ` swp - Re: ext3 journal on software raid (was Re: PROBLEM: Kernel 2.6.10 crashing repeatedly and hard) Mark Hahn
2005-01-09 22:20                                       ` Alvin Oga
2005-01-06  5:01                                 ` Alvin Oga
2005-01-05 17:07                     ` Guy
2005-01-05 17:21                       ` Alvin Oga
2005-01-05 17:32                         ` Guy
2005-01-05 18:37                           ` Alvin Oga
2005-01-05 17:34                         ` ECC: RE: ext3 blah blah blah Gordon Henderson
2005-01-05 18:33                           ` Alvin Oga
2005-01-05 17:26                       ` ext3 journal on software raid (was Re: PROBLEM: Kernel 2.6.10 crashing repeatedly and hard) David Greaves
2005-01-05 18:16                         ` Peter T. Breuer
2005-01-05 18:28                           ` Guy
2005-01-05 18:26                         ` Guy
2005-01-05 15:48                   ` Peter T. Breuer
     [not found]       ` <41D45C1F.5030307-XAri/EZa3C4vJsYlp49lxw@public.gmane.org>
2004-12-30 20:54         ` PROBLEM: Kernel 2.6.10 crashing repeatedly and hard berk walker
2004-12-30 20:54           ` berk walker
2005-01-01 13:39         ` Helge Hafting
2005-01-01 13:39           ` Helge Hafting
2005-01-07  6:21     ` Clemens Schwaighofer
2005-01-07  9:39       ` Andy Smith
2004-12-30 20:40 ` Linus Torvalds
2004-12-30 22:04   ` Georg C. F. Greve
2004-12-30 22:23     ` Linus Torvalds
2004-12-31  4:16       ` Herbert Poetzl
2004-12-31  4:53         ` Linus Torvalds
2004-12-31 10:55           ` Russell King
2004-12-31  9:58       ` Georg C. F. Greve
2005-01-01 20:06       ` Alan Cox
     [not found]     ` <m3zmzvl9x1.fsf-glUV91rXKAHWIjgkaejU9x2eb7JE58TQ@public.gmane.org>
2005-01-02 20:11       ` Linus Torvalds
2005-01-02 20:11         ` Linus Torvalds
     [not found]         ` <Pine.LNX.4.58.0501021147260.2280-gWtpgVMusWVb5UGfqNBoRg@public.gmane.org>
2005-01-02 20:32           ` Dave Jones
2005-01-02 20:32             ` Dave Jones

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=41DC9420.5030701@h3c.com \
    --to=mhardy@h3c.com \
    --cc=linux-raid@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.