Poor man's RAID?

Poor man's RAID?

Post by Christopher Brow » Sun, 19 Apr 1998 04:00:00




>Hello,

>I have a couple of new ATA IDE drives, one per channel and I would
>like to optimize the partitioning (for speed).

>The drives are closely matched in performance and both have ~1gig free
>at the front.

>Would I see any performance gain if I split the distribution (Stampede
>linux) over, say, 30 small partitions, 15 per drive? eg:

There are two benefits that you can get:

a) If (for instance) partition /var gets hit a lot, and partition /usr
gets hit a lot, by putting them on separate disks, requests can be
simultaneously serviced.

The main way expanding to 15 (or 30) partitions would help you is if you
actually had 15 disks, and thus could "multitask" the drive controllers.
Chances are that there are only a couple of partitions that are worth
separating in this fashion.

Approach: Figure out which two partitions are accessed most, and make
sure they're on different drives.

Chances are that this only means moving around one partition, or two at
most.

b) Then there's striping, which sounds like it's what you're actually
thinking of.

In this case, there is a single partition that you would like to allow
to use both drives at once, pulling successive clusters from both
disks.

Here, you should figure out which single partition is accessed most, and
create two partitions (one on each disk) join them using MD, and put an
ext2fs onto the joined partition.

You could extend this to multiple partitions.  Thus, if /var and /usr
get "hit a lot", you could put 4 partitions onto the two disks, and
have part of /var be on each disk, and part of /usr be on each disk.

Be aware that this substantially reduces reliability, because if either
disk dies, the partition dies.

Rule #1: Make sure / is *NOT* striped.

If either disk dies, and / is striped, then your system WILL NOT BOOT.
That is a very bad thing.

Rule #2: More usefully, make sure that you have a bootable system even
if all of the striped partitions get toasted.

Partitions that are so highly active that they are worth striping are
generally running some particular application.  Things that are
typically so I/O intense include:

- News (INN/CNEWS)

- Database applications.  In expensive commercial realms, things like
Oracle/Sybase/Informix commonly use striping.  More likely, under Linux,
you are running PostgreSQL, mySQL, SOLID, Adabas-D, or something else
listed at <http://www.hex.net/~cbbrowne/rdbms.html>.

If you haven't done enough benchmarking or other testing to establish
that striping a particular partition will result in significant
performance improvements, then chances are that it will have negligible
effects on performance, and merely make your system more fragile,
substantially reducing your system's MTBF.

In any case, if you can't establish pretty clearly that the change will
significantly improve performance (your "tangible benefit"), don't
bother optimizing it.

Michael Jackson's First Rule of Software Optimization: "Don't Do It."

Michael Jackson's Second Rule of Software Optimization: "Don't Do It...
Yet..."

(No joke intended here; I am not referring to "MJ of the gloved hand,"
but rather of MJ of Computer Science/Software Engineering fame...)

--
"The surest sign that intelligent life exists elsewhere in the universe
is that it has never tried to contact us." -- Calvin and Hobbes

 
 
 

Poor man's RAID?

Post by < » Mon, 20 Apr 1998 04:00:00


Having 2 swap partitions split across the 2 hard drives could potentially
be a big advantage if you run at the limits of your memory...

 
 
 

Poor man's RAID?

Post by Jason Cliffor » Thu, 23 Apr 1998 04:00:00



> Having 2 swap partitions split across the 2 hard drives could potentially
> be a big advantage if you run at the limits of your memory...

but with memory prices as low as they are now why not just upgrade?
--
Jason Clifford
Genesis Linux Systems
http://genesis.ukpost.com
 
 
 

Poor man's RAID?

Post by Jim Buchan » Thu, 23 Apr 1998 04:00:00


: >
: > Having 2 swap partitions split across the 2 hard drives could potentially
: > be a big advantage if you run at the limits of your memory...
:
: but with memory prices as low as they are now why not just upgrade?

Well, you could be like me. I have 64M in my favorite Linux machine,
and rarely use more than about 36M of it. But every once in a while
I'll be debugging a program with a memory diagnostic library, such as
Bruce Peren's Electric Fence (recommended), and I'll need 256M or more
of memory. Hardly smart to upgrade for something like that when it
only happens a few times a year.

--

================= http://members.iquest.net/~jbuchana ====================
"ID was everything nowadays. Used to be that money was gold or silver or
 something tangible. But plastic money was just a way of telling people
 where to find you. Who you were. How to touch you." -Bruce Sterling
==================== http://hybiss.delcoelect.com ========================

 
 
 

Poor man's RAID?

Post by Shaun Cas » Tue, 28 Apr 1998 04:00:00


What's all this got to do with RAID?  Hmm.. Iguess you guys are not well
informed on the subject, huh?



> > Having 2 swap partitions split across the 2 hard drives could potentially
> > be a big advantage if you run at the limits of your memory...

> but with memory prices as low as they are now why not just upgrade?
> --
> Jason Clifford
> Genesis Linux Systems
> http://genesis.ukpost.com

 
 
 

Poor man's RAID?

Post by that » Sat, 02 May 1998 04:00:00



> What's all this got to do with RAID?  Hmm.. Iguess you guys are > not well informed on the subject, huh?

Perhaps they might like to stripe them using md a la RAID0??


>>> Having 2 swap partitions split across the 2 hard drives could >>> potentially be a big advantage if you run at the limits of >>> your memory...

>> but with memory prices as low as they are now why not just
>> upgrade?

 
 
 

1. Poor man's RAID?

Hello,

I have a couple of new ATA IDE drives, one per channel and I would
like to  optimize the partitioning (for speed).

The drives are closely matched in performance and both have ~1gig free
at the front.

Would I see any performance gain if I split the distribution (Stampede
linux) over, say, 30 small partitions, 15 per drive? eg:

/dev/hda        /dev/hdc
--------        --------
/bin            /sbin
/usr/sbin       /usr/bin
/dev            /var
...             ...

I don't know enough about ext2fs to know if there is any performance
hit in continually moving across partitions or drives?

I'm not all that concerned about admin hassles or wasted space but I
don't want to go this route unless there is some tangible benefit...

Patrick
__________

"It's better to have loved and lost than to never have seen Lost in Space"  
    -- Kelly Bundy quoting Chinese philosopher Unconcious

2. Why does my post disapeare after 2 days, no answer ?

3. Poor man's LAN?

4. login problem

5. Poor man's Unix?

6. aic7xxx still broken in 2.0.34 and 2.1.107

7. netmask=0.0.0.1 (poor man's load balancing)

8. PPP problem when not root

9. loading executables into the data segment (poor man's exec)

10. Poor man's Unix?

11. New FreeBSD Tutorial - A Poor Man's Web Server!

12. mpr-1.5 - a poor man's memory profiler

13. Poor man's high availability.