Please help me fill in the blanks.
Hi again.
I am doing a thesis about comparing the Linux kernel against HP-UX, AIX, Tru64 UNIX, and Solaris (as you probably alredy know).
I'm stuck now (and the thesis has to bee ready until tomorow) with some few features that the other operating system have, and I can't find any information about it, if those features are implemented in Linux or not.
The features that I'm wondering about are:
* Dynamic Processor Resilience
* Dynamic Memory Resilience
* Dynamic Page Sizing
* Live Upgrade
* Alternative I/O Pathing
* HSM
* TCP selective acknowledgement (SACK)
* Service Location Protocol (SLP)
* ATM IP switching
* SOCKS 5 support
* Multilink PPP
* TCP/IP Gratuitous ARP (RFC 2002)
* Path MTU Discovery (RFC 1191)
* Path MTU Discovery over UDP
* IP Multipath Routing
The questions I have about the features above are:
* Are any of the above features implemented in the kernel? If yes, where can I read (url-link to the article, paper... please) about it?
* Is any of the features implemented through a program /daemon? If yes, which program (link to program)?
Here is the link to my thesis if anyone hasn't got my previous message:
http://www.student.hig.se/~na98csa/{
,Linux.tex,xjobb.ps}
Thanks in advance,
Cesar da Silva
_____________________________________________________
Do You Yahoo!?
[email protected] - skaffa en gratis mailadress på http://mail.yahoo.se
Quoted 2 lines · Cesar Da Silva
The features that I'm wondering about are:
* Dynamic Processor Resilience
is this fault tolerance? I think if a CPU croaks, you are dead.
There are patches for hot swap cpu support, but I haven't seen any CPU fault tolerance patches that can handle a dead processor
Quoted 1 line
* Dynamic Memory Resilience
RAM fault tolerance? There was a patch a long time ago which detected bad ram, and would mark those memory clusters as unuseable at boot. However that is clearly not dynamic.
If your memory croaks, your kernel will experience random corruptions
Quoted 1 line
* Dynamic Page Sizing
no
Quoted 1 line
* Live Upgrade
LOBOS will let one Linux kernel boot another, but that requires a boot step, so it is not a live upgrade. so, no, afaik
Quoted 1 line
* Alternative I/O Pathing
be less vague
Quoted 1 line
* HSM
patches exist, I believe
Quoted 1 line
* TCP selective acknowledgement (SACK)
yes
Quoted 1 line
* Service Location Protocol (SLP)
don't know
Quoted 1 line
* ATM IP switching
yes, I believe
Quoted 1 line
* SOCKS 5 support
yes, via userspace apps/libs
Quoted 1 line
* Multilink PPP
yes
Quoted 1 line
* TCP/IP Gratuitous ARP (RFC 2002)
not sure
Quoted 1 line
* Path MTU Discovery (RFC 1191)
yes
Quoted 1 line
* Path MTU Discovery over UDP
not sure, but I think so
Quoted 1 line
* IP Multipath Routing
yes
Quoted 4 lines
The questions I have about the features above are:
* Are any of the above features implemented in the kernel? If yes, where can I read (url-link to the article, paper... please) about it?
Signature · 3 lines
Jeff Garzik | Disbelief, that's why you fail. Building 1024 | MandrakeSoft |
Quoted 4 lines
Quoted 1 line
* Live Upgrade
LOBOS will let one Linux kernel boot another, but that requires a boot step, so it is not a live upgrade. so, no, afaik
If you build nearly everything (except, obviously what you need to boot) as modules, you can unload modules, build new versions, and reload them. So, you could say that partial support for "live upgrades" is included.
It works, too - I unloaded my OV511 driver a few weeks ago, copied the source for the new one in, built it, and re-inserted it. Same goes for the DRM module a couple of weeks before that. Now, the machine in question gets rebooted fairly often in any case, but those were things I *didn't* have to reboot for.
--------------------------------------------------------------
from: Jonathan "Chromatix" Morton mail: [email protected] (not for attachments) big-mail: [email protected] uni-mail: [email protected]
The key to knowledge is not to rely on people to teach you it.
Get VNC Server for Macintosh from http://www.chromatix.uklinux.net/vnc/
-----BEGIN GEEK CODE BLOCK-----
Version 3.12
GCS$/E/S dpu(!) s:- a20 C+++ UL++ P L+++ E W+ N- o? K? w--- O-- M++$ V? PS PE- Y+ PGP++ t- 5- X- R !tv b++ DI+++ D G e+ h+ r++ y+(*) -----END GEEK CODE BLOCK-----
Quoted 8 lines · Jeff Garzik
Quoted 2 lines · Cesar Da Silva
The features that I'm wondering about are:
* Dynamic Processor Resilienceis this fault tolerance? I think if a CPU croaks, you are dead.
There are patches for hot swap cpu support, but I haven't seen any CPU fault tolerance patches that can handle a dead processor
The S/390 has this; presumably it applies to Linux as well as the other supported OSs?
Quoted 7 lines
Quoted 1 line
* Dynamic Memory Resilience
RAM fault tolerance? There was a patch a long time ago which detected bad ram, and would mark those memory clusters as unuseable at boot. However that is clearly not dynamic.
If your memory croaks, your kernel will experience random corruptions
ECC can be supported by the hardware; no support for mapping out duff banks on x86, but again S/390 may differ?
Quoted 4 lines
Quoted 1 line
* Live Upgrade
LOBOS will let one Linux kernel boot another, but that requires a boot step, so it is not a live upgrade. so, no, afaik
Live SOFTWARE upgrade, or live HARDWARE upgrade? If the latter, things like hotswap PCI, USB... and again the S/390?
Quoted 3 lines
Quoted 1 line
* Service Location Protocol (SLP)
don't know
Yes, I think so - mars_nwe surely needs this?
Quoted 3 lines
Quoted 1 line
* TCP/IP Gratuitous ARP (RFC 2002)
not sure
Isn't that how LVS clusters handle IP takeovers?
Quoted 3 lines
Quoted 1 line
* Path MTU Discovery (RFC 1191)
yes
With one or two RFC violations, yes.
Basically, most of those features relating to hardware resilience should be usable with Linux on an S/390 - they are hardware features, though, AFAICS?
James.
Thank you Jeff for your very helpfull answer.
--- Jeff Garzik <[email protected]> skrev: >
Quoted 3 lines · Cesar Da Silva
Quoted 1 line
* Alternative I/O Pathing
be less vague
What I mean with the above (my defenition) is:
[Alternative I/O Pathing allows the operating system to re-route the I/O of devices, such as disk or network adapters, to a backup device, in case of failure.
Regards,
Cesar da Silva
_____________________________________________________
Do You Yahoo!?
[email protected] - skaffa en gratis mailadress på http://mail.yahoo.se
Quoted 9 lines · Jonathan Morton
Quoted 5 lines
Quoted 2 lines
* Live Upgrade
LOBOS will let one Linux kernel boot another, but that requires a boot step, so it is not a live upgrade. so, no, afaik
If you build nearly everything (except, obviously what you need to boot) as modules, you can unload modules, build new versions, and reload them. So, you could say that partial support for "live upgrades" is included.
I stand corrected, though I clearly know better:
Modules are unloaded/reloaded all the time during my driver development :)
Signature · 3 lines
Jeff Garzik | Disbelief, that's why you fail. Building 1024 | MandrakeSoft |
Quoted 1 line · Jeff Garzik
Quoted 1 line
* Service Location Protocol (SLP)
www.openslp.org
Regards
Ingo Oeser
Signature · 2 lines
To the systems programmer, users and applications serve only to provide a test load.
Quoted 4 lines · Jeff Garzik
Quoted 2 lines
* Service Location Protocol (SLP)
don't know
Userspace: http://www.openslp.org/
Quoted 3 lines
Quoted 1 line
* TCP/IP Gratuitous ARP (RFC 2002)
not sure
Userspace. Also no tool comes to my mind, arping should be easily modified to do this.
Dominik
Signature · 3 lines
A lovely thing to see: Kobayashi Issa
through the paper window's holes (1763-1828)
the galaxy. [taken from: David Brin - Sundiver]Quoted 5 lines · On Sun, 27 May 2001, [iso-8859-1] Cesar Da Silva wrote:
I am doing a thesis about comparing the Linux kernel against HP-UX, AIX, Tru64 UNIX, and Solaris (as you probably alredy know).
I'm stuck now (and the thesis has to bee ready until tomorow)
Aren't you the same guy who posted this question last week?
And the same guy who asked us to "fill in the blanks" on an essentially EMPTY work? ;)
Rik
--
Virtual memory is like a game you can't win; However, without VM there's truly nothing to lose...
http://www.surriel.com/ http://distro.conectiva.com/
Send all your spam to [email protected] (spam digging piggy)
--- Rik van Riel <[email protected]> skrev: > On
Sun, 27 May 2001, [iso-8859-1] Cesar Da Silva
Quoted 11 lines
Quoted 1 line · wrote:
I am doing a thesis about comparing the Linux
kernel
Quoted 1 line
against HP-UX, AIX, Tru64 UNIX, and Solaris (as
you
Quoted 2 lines
probably alredy know).
I'm stuck now (and the thesis has to bee readyuntil
Quoted 1 line
tomorow)
Quoted 7 lines
Aren't you the same guy who posted this question
last
week?And the same guy who asked us to "fill in the
blanks"
on an essentially EMPTY work? ;)
Yes, but this time I was more precise with what I wanted help with (I included a list in the message), and it helped, I got alot more responses this time.
Regards,
Cesar da Silva
_____________________________________________________
Do You Yahoo!?
[email protected] - skaffa en gratis mailadress på http://mail.yahoo.se
Quoted 5 lines
Quoted 1 line
* Dynamic Memory Resilience
RAM fault tolerance? There was a patch a long time ago which detected bad ram, and would mark those memory clusters as unuseable at boot. However that is clearly not dynamic.
If you are referring to Badram patch by Rick van Rein
(http://rick.vanrein.org/linux/badram/), it doesn't detect the bad ram, memtest86 does that part (and does it well) -- you enter then enter the badram clusters as boot param. But I have to say badram patch works marvellously (thanks, Rick.) Shame it didn't find its way to standard kernel.
-- v --
--- Ville Herva <[email protected]>
skrev: > > > * Dynamic Memory Resilience
Quoted 7 lines
Quoted 2 lines
RAM fault tolerance? There was a patch a long
time ago which detected
Quoted 1 line
bad ram, and would mark those memory clusters as
unuseable at boot.
Quoted 1 line
However that is clearly not dynamic.
Quoted 11 lines
If you are referring to Badram patch by Rick van Rein
(http://rick.vanrein.org/linux/badram/), it doesn't detect the bad ram,
memtest86 does that part (and does it well) -- you enter then enter the
badram clusters as boot param. But I have to say badram patch works
marvellously (thanks, Rick.) Shame it didn't find its way to standard
kernel.
Hi Ville, and thanks for the great link and the
information.
Regards,
Cesar da Silva
_____________________________________________________
Do You Yahoo!?
[email protected] - skaffa en gratis mailadress på http://mail.yahoo.se