I found an article that I believe was done by Ars and they tested a database load on several SSD drives and said the reality of the longevity of the SSD vs. the HDD was non-existant and the best estimate cases showed an SSD actually lasting LONGER than the MTBF of a regular hard drive. So I am not convinced that these drives fail so quickly. The first 5 or 6 generations did, but I believe the newer systems will last just as long as a HDD under all similar loads. I will have to look for that article again and post it.

Nathan

On Sun, Mar 7, 2010 at 2:59 PM, Stephen <cryptworks@gmail.com> wrote:
I have done some real work on SSD and their performance is great Seek
times are non existent, throughput is about as good as the interface
gets.

but note if you do allot of disk read write they will fail sooner than
a regular HDD it is simply a wearing factor of the media.

The algorithms and the like built in are very good so you will likely
get a decent life from the drive.

the only 2 tests i have done on them personally were Web IO and DB IO
and they were about 10x as fast as a SATA 2 hdd overall (2 drive
mirror on a highpoint 3210)

http://www.highpoint-tech.com/USA/rr3120.htm

I find Raid 0 gained me more performance with less money (or raid 10
if you want the redundancy) over a single drive.


On Fri, Mar 5, 2010 at 5:47 PM, Nathan England <nathan@paysonlinux.org> wrote:
> I have read a dozen sites discussing various benchmarkings between myriad
> setup of SSD's vs. HDD's. I have not seen any code monkey reviews. Has any
> one purchased one of these for use in compiling code, or maybe you're a
> gentoo fan???
> I compile a ton of stuff everyday and I'm curious if it is worth the money
> to try a SSD. I also have several large database systems I would like to
> migrate to a SSD, but I have not looked for reviews with databases yet.
> Nathan
> ---------------------------------------------------
> PLUG-discuss mailing list - PLUG-discuss@lists.plug.phoenix.az.us
> To subscribe, unsubscribe, or to change your mail settings:
> http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss
>



--
A mouse trap, placed on top of your alarm clock, will prevent you from
rolling over and going back to sleep after you hit the snooze button.

Stephen
---------------------------------------------------
PLUG-discuss mailing list - PLUG-discuss@lists.plug.phoenix.az.us
To subscribe, unsubscribe, or to change your mail settings:
http://lists.PLUG.phoenix.az.us/mailman/listinfo/plug-discuss