Re: A7N8X-e RAID 0 Stripe Size advice??

I just happened to be in the right place at the right time to see what
Tim wrote in alt.comp.periphs.mainboard.asus and:

> when it comes to stripe size there is only one definitive answer:
> benchmark your applications to find out what is best for your
> circumstances. To facilitate this, I would install the OS on its own
> HDD and trial the different stripes using RAID 0 for data only - this
> is a sensible config anyway as you no doubt appreciate that loss of 1
> drive in RAID 0 = loss of all partitions on both drives - this is
> considerably more probable than the loss of one disc.
>
> For Image Editing I would expect a clear outcome - no idea what
> though!
>
> - Tim
>
>
> "dogbowl" <dogbowl@xxxxxxxxxxxx> wrote in message
> news:Xns96D2BDE142F2Bdogbowldeluxe@xxxxxxxxxxxxxxxxxx
>> Hi group...
>>
>> Just for kicks I'm about to put together a 2 disc RAID 0 array on my
>> A7N8X-
>> e... It's the first one I've built so I thought I'd pay a visit to
>> the experts who've helped me out a few times before ;-)
>>
>> I've got my discs (2 x 300Gb Maxtor Maxline 3's), got my drivers (v
>> 1.0.50), been using mobo BIOS 1013 for some time, got almost
>> everything I need but would anyone in here care to recommend a stripe
>> size...?
>>
>> I mainly mess about with MP3's, DVD, Image Editing... and the OS will
>> be on
>> the same array
>>
>> Anyone care to throw in their tuppence?
>>
>> Any replies appreciated ;-)
>>
>> dogbowl

Thanks for the response Tim... but I'm already up and running... the
install went super smoothly and the machine is fair zipping along
compared to yesterday when it was plain ol' IDE... took a fair while to
format the 600Gb array though :-(

OS, Apps and Data all on the same array... Not ideal but I'm fairly
sensible about backing up data and the OS / Apps are not that big a deal
to re-install in the event of a catastrophe...

I chose the 64Kb stripe for this, my first build, and the performance
increase has been dramatic... I suspect the re-install of XP-Pro might
have also contributed to the performance gains... the machine hadn't had
a rebuild for over a year.

I'll keep an eye on it and if I get fed up I'll start again from
scratch. Thanks for the tips

dogbowl



.

Relevant Pages

  • Re: SBS2003 rebuild
    ... I agree he should just image the server using an appropriate WinPE ... disk, wipe the array, create and initialize a new array (and allow the ... ServeRAID to completely scrub the drives), ... A bad stripe isn't the result of a single drive failure....something ...
    (microsoft.public.windows.server.sbs)
  • Re: Multiple LUNs vrs one large stripe. Opinions?
    ... but I'd need Veritas to put that IN WRITING ... > HW stripe, plus a min of 4 members in each SW stripe to yield benefits ... Setting a particular size lun on a back end array for a particular job ...
    (comp.unix.solaris)
  • Re: Full restore - with time to plan it
    ... assuming now that I will have backed up the bad data in the bad stripe. ... of the info given by Super, it appears that an image is unlikey to be ... BTW, I did Google for the array stripe failure condition, but had so ... Replacing drives in the array does not necessarily clear the condition ...
    (microsoft.public.windows.server.sbs)
  • Re: [PATCH 000 of 5] md: Introduction
    ... separate userspace program which operates on an inactive array, with an ability to plan it's I/O strategy in alot more efficient and safer way... ... I've been kicking around the idea of a user space utility that manipulates the device mapper tables and performs block moves itself to reshape a raid array. ... what the raid will look like when done, i.e. stripe of sda, sdb, and sdc ... Initially the length of the chunks from raid-progress and raid-new are zero, so you will still be entirely accessing raid-old. ...
    (Linux-Kernel)
  • Re: SBS2003 rebuild
    ... I agree he should just image the server using an appropriate WinPE ... disk, wipe the array, create and initialize a new array (and allow the ... ServeRAID to completely scrub the drives), ... A bad stripe isn't the result of a single drive failure....something ...
    (microsoft.public.windows.server.sbs)