HDD Live Install ...... Poorman's Install

[Just18]

Thanks for testing it Vorteks 

[Vortеx]

Thanks for testing it Vorteks 

I wouldn't say "testing". I would say "enjoying"

[melodie]

Hi,

Where is this project now ? Just19, are you going to continue working on this one ? I didn't have too much time for getting involved this summer, but I still follow and have interest for this tread.

Regards,
Mélodie

[Just18]

Hi,

Where is this project now ? Just19, are you going to continue working on this one ? I didn't have too much time for getting involved this summer, but I still follow and have interest for this tread.

Regards,
Mélodie

It reached a Beta 3 stage and got left at that.
I have not worked on it for quite some time.

Seems I rather ran out of steam as I have not updated the LiveUSB script either, although I have worked on changes to it.

The third script in the series  ISO2USB  was also started and got to a working stage ....  and also not finished.

I am presently in the middle of reconstruction works at home and am not very inclined to get into anything for a couple of more weeks.

Maybe in a week or two, if there is much interest, I can revive the scripts and make them available generally, after further testing.

....  time will tell ....   

[mimas]

This program sounds very nice. If the "installation" is made on the hard drive maybe it's time to look toward a fantastic compression algorithm. It's lightning fast, it requires no memory for decompression and has the best ratio compression/speed. Its name is LZO. You should read the doc, my english is too rusty to explain LZO's interests correctly.

XZ is a very good compressor but good compression is not needed if the squashfs root is in the harddrive. In my humble opinion, the most important part here is speed: the system should be as fast as a normal installation. LZO makes that possible.

But LZO is not active in PClinuxOS squashfs tools so a new package should be made (a line to uncomment in a Makefile).

LZO should make PClinuxOS ISO to run faster if, of course, the LZOed squashfs fits in a CDROM.

[Just18]

This program sounds very nice. If the "installation" is made on the hard drive maybe it's time to look toward a fantastic compression algorithm. It's lightning fast, it requires no memory for decompression and has the best ratio compression/speed. Its name is LZO. You should read the doc, my english is too rusty to explain LZO's interests correctly.

XZ is a very good compressor but good compression is not needed if the squashfs root is in the harddrive. In my humble opinion, the most important part here is speed: the system should be as fast as a normal installation. LZO makes that possible.

But LZO is not active in PClinuxOS squashfs tools so a new package should be made (a line to uncomment in a Makefile).

LZO should make PClinuxOS ISO to run faster if, of course, the LZOed squashfs fits in a CDROM.

It is available in the repository ......  as  lzop  package ....

LZO fast file compressor
lzop is a file compressor similar to gzip. Its main advantages over gzip
are much higher compression and decompression speed at the cost
of compression ratio.

To make use of it, it would have to be incorporated into the mylivecd utility, I think.

For LiveHDD install the squash file does not have to fit on a CD .....  DVD size is definitely available .....  and maybe with some edits the ISO limit of 4GBs could be eliminated too.

First of course it would have to be available for use when creating a remaster 

.

[melodie]

But LZO is not active in PClinuxOS squashfs tools so a new package should be made (a line to uncomment in a Makefile).

First of course it would have to be available for use when creating a remaster 

.

In mylivecd sure, to get and added option, such as  "--xz" and "--gzip", and as I understand it, have it's option activated when compiling squashfs tools...

[AS]

But LZO is not active in PClinuxOS squashfs tools so a new package should be made (a line to uncomment in a Makefile).

First of course it would have to be available for use when creating a remaster 

.

In mylivecd sure, to get and added option, such as  "--xz" and "--gzip", and as I understand it, have it's option activated when compiling squashfs tools...

+1,
first add LZO support in mksquashfs, second add support in mylivecd, third add support in poormanHDD & LiveUSB-Creator 

[Just18]

Yes of course ....  forgot about support in mksquashfs ...  had not checked ... 

From mksquashfs Help ....

Filesystem build options:
-comp <comp>            select <comp> compression
                        Compressors available:
                                gzip (default)
                                lzma
                                xz

[mimas]

> first add LZO support in mksquashfs, second add support in mylivecd, third add support in poormanHDD & LiveUSB-Creator

PoormanHDD doesn't need LZO support as it should be transparent. Nothing looks like more squashfs than a squashfs file.

I was just throwing an idea in the air. I used "LiveHDD" around 2007 and It was not possible to get a very fast decompressor for the rootfs at this time. The computer was running fine, though. Public access computers running PClinuxOS with a "liveHDD" (easy to manage with cp, no direct modifications possible for users) and a lzo compressed filesystem for speed. I can't get it out of my head. :p

This is just a detail. In fact this idea is not really PoormanHDD related, it's more about the concept

[AS]

A comparison between LZO/LZMA/GZIP:

http://free-electrons.com/blog/lzo-kernel-compression/

[melodie]

A comparison between LZO/LZMA/GZIP:

http://free-electrons.com/blog/lzo-kernel-compression/

Hi,

What is the percentage in the third row (So, here are the figures, average on 20 boots with each compression method) representing ? It seems according to this frame that gzip could be the fastest to uncompress while lzo would compress less and be a little slower ? Maybe I don't understand it correctly ?

[AS]

A comparison between LZO/LZMA/GZIP:

http://free-electrons.com/blog/lzo-kernel-compression/

Hi,

What is the percentage in the third row (So, here are the figures, average on 20 boots with each compression method) representing ? It seems according to this frame that gzip could be the fastest to uncompress while lzo would compress less and be a little slower ? Maybe I don't understand it correctly ?

Hi Melodie,

probably you mean the 3rd column, that look like the percentage of time to boot compared to the gzip compressed kernel.
LZO look like being a little faster to boot, while compressing a little less.

The table is not very clear 

AS

[melodie]

Hi AS,

So reading the end of the page:

LZO looks like a good candidate when it comes to speeding up the boot process, at the expense of some (almost neglectable) extra space. Gzip is close to LZO when it comes to size, although extraction is not as fast. That means that unless you’re hitting corner cases, like only having enough space for a Gzip compressed image but not for one made with LZO, choosing the latter is probably a safe bet.

Besides, the LZO-compressed kernel size is about 54% the size of the uncompressed kernel. As the kernel load time varies linearly with its size, load time for an uncompressed kernel doubles. While 0.55s are won because there’s no need to run a decompression algorithm, you spend twice as much time loading the kernel. This time is not negligible at all compared to the decompression time. Indeed, loading the uncompressed image takes roughly 0.8s. That means that at the cost of slowing down the boot process by 0.15s (compared to an uncompressed kernel), one gets a kernel image which is roughly twice as small. Rather nice, isn’t it?

This page seems to confirm mimas's suggest when saying it would be nice to have lzo as one more build option with our mksquashfs ?

Code:

# mksquashfs --help
SYNTAX:mksquashfs source1 source2 ...  dest [options] [-e list of exclude
dirs/files]

Filesystem build options:
-comp <comp> select <comp> compression
Compressors available:
gzip (default)
lzma
xz
.....

[AS]

This page seems to confirm mimas's suggest when saying it would be nice to have lzo as one more build option with our mksquashfs ?

It could be interesting to give a try, however if the line to enable LZO is commented out probably there is a reason ... that I don't know.

AS