All
Thank you for the responses.
When I wrote the initial email I left out detail in order to keep the things
short.
To address a few of the concerns and suggestions.
The construction of the Disk Farm will be as needed starting
with two 200G dirves raid 1 that will be NFS and Samba advertised
and two 160G drives raid 1 that will be separately advertised.
I picked these up yesterday. In the future raid 1 pairs will be
added as required and offered as individual resources that I will manage.
Other members of my family tend to use disk space and I need to control
whos computer can consume what. I am by far the biggest culprit.
One other point, that I will test, is in the event of a server or
disk failure I can disconnect the good drive and connect it to
another system and manually mount it to retrieve the data
instead of replacing the failed drive and rebuilding the raid.
This is why I am using raid 1 and not raid 5.
ANY COMMENT ON THIS appreciated.
160G ATA drives can be found around here for $39.99 on
sale and $200G for $89 any day of the week.
Each drive in a pair will be connected through separate
USB ports on the server and USB Hubs. This pushes
the single point of failure back to the server on one end
the power supply in the disk farm on the other.
I am using USB to IDE cables without individual power
supplies. All of the drives will be mounted in a 2U case
with a PC Power and Cooling supply with an opto-isolater
controlling the supply turn on slaved to the servers 5 volt
supply. Unused outputs on the PC Power and Cooling
supply that require a minimum load will be terminated with
resistors. As I add disks the spin up load on the supply
will be measured to stay in spec. The nice thing with
the USB connection is I can start another farm in another
case, within reason. The number 14 for total disk count
is limited by the case size and more importantly the
thermal management inside the case. A serial connected
temperature probe located near the air exit port will be
monitored by the server.
Professionally I use Sun Solaris on high end machines
and am switching to RedHat from Unixware for the low end.
I run Oracle on both. Personally I use the CENTOS, which
is where the disk farm will be attached. A second CENTOS
system will take over by Oracle database from Unixware.
I picked CENTOS because I only have time to port Unixware
to one other OS and CENTOS is RedHat. RedHat was
selected purely based on support availability for the OS/Oracle combination.
Further suggestions or discussion are welcome
Bill Hess
[EMAIL PROTECTED]
On Thu Jan 12 12:20 , [EMAIL PROTECTED] (Ross Vandegrift) sent:
On Thu, Jan 12, 2006 at 11:16:36AM +0000, David Greaves wrote:
> ok, first off: a 14 device raid1 is 14 times more likely to lose *all*
> your data than a single device.
No, this is completely incorrect. Let A denote the event that a single
disk has failed, A_i denote the event that i disks have failed.
Suppose P(A) = x. Then by Bayes's Law the probability that an n disk RAID
will lose all of your data is:
n_1 = P(A) = x
n_2 = P(A_2) = P(A) * P(A_1 | A) = x^2
n_3 = P(A_3) = P(A) * P(A_2 | A) = x^3
...
n_i = P(A_i) = P(A) * P(A_{i-1} | A) = x^i
ie, RAID1 is expoentially more reliable as you add extra disks!
This assumes that disk failures are independant - ie, that you
correctly configure disks (don't use master and slave on an IDE
channel!), and replace failed disks as soon as they fail.
This is why adding more disks to a RAID1 is rare - x^2 is going to be
a really low probability! It will be far, far more common for
operator error to break a RAID than for both devices to honestly fail.
--
Ross Vandegrift
[EMAIL PROTECTED]
"The good Christian should beware of mathematicians, and all those who
make empty prophecies. The danger already exists that the mathematicians
have made a covenant with the devil to darken the spirit and to confine
man in the bonds of Hell."
--St. Augustine, De Genesi ad Litteram, Book II, xviii, 37
[EMAIL PROTECTED] (Ross Vandegrift)
David Greaves <[EMAIL PROTECTED]>
[EMAIL PROTECTED], [email protected]
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
