AW: RAID1 VS RAID5

AW: RAID1 VS RAID5

[Martin Bene]

>> OTOH, you have access to the source code if you really want 
>> to try it 
>> another way.
>
>So again my first question thats seems to go ignored: does someone have
>made a patch? Is there some line to put in raidtab to change raid1
>behavior?

There used to be a patch for that, from the readme:

> Operation:

> In normal raid1 code (raidtools 0.90), reads are distributed
> in a very fine grained fashion, new read will get new disk.
> Only exception are sequential reads, which are never distributed.

> This patch records a logical linear position of head in every
> disk in array. Then when read requests is made it tries to
> find a disk which head is closest to the requested block and
> uses that disk. Initial measurements show performance gain
> in reading and in seeking with many readers.

> Flames, comments, improvements, benchmarks greatly 
> appreciated. Send them to miku@iki.fi

No idea if this is still applicable to current raid code or still maintained;
the old stuff used to be at 

http://www.icon.fi/~mak8/raid1/

Yours, martin

Re: AW: RAID1 VS RAID5

[Mario Giammarco]

Il mar, 2003-10-28 alle 11:51, Martin Bene ha scritto:

> >So again my first question thats seems to go ignored: does someone have
> >made a patch? Is there some line to put in raidtab to change raid1
> >behavior?
> 
> There used to be a patch for that, from the readme:


> No idea if this is still applicable to current raid code or still maintained;
> the old stuff used to be at 
> 
> http://www.icon.fi/~mak8/raid1/

Thank you, I remember about a patch. This seems very old I will try to
email the maintainer.
-- 
Mario Giammarco