Scan are doing 40gb SSD Intel X25's for £84
http://www.scan.co.uk/product.aspx?ProductId=34904
(today only price a bit cheaper than that)
Is it worth RAID'ing two of those instead of buying one
80gb for same price?
Printable View
Scan are doing 40gb SSD Intel X25's for £84
http://www.scan.co.uk/product.aspx?ProductId=34904
(today only price a bit cheaper than that)
Is it worth RAID'ing two of those instead of buying one
80gb for same price?
They're faster than the 80GB but you lose TRIM support. Although i didn't get a reply in another thread, maybe the effects of not having TRIM aren't that bad as only half of the data is being written to each drive in RAID 0. Just a thought.
I'm confused, I did read article, is this entirely due to the limitations of a single SATA port?
Because it was my understanding that the way these drives work they are like a little RAID inside, so the 40GB had half the channels stripped out, and therefore would be half as fast,
Likewise the 160GB drives got a speed bump with the firmware upgrad because they have more channels to address (probably not using right terminology), like they are two 80GB drives in RAID.
I wouldve expected them to perform equally to an X25-M, but not better :S
I have 2 40gb intels in raid (Kingston's actually), it's better than the single as you get same write speeds but more read, which is a huge gain.
Also, you can have trim as well:
http://www.bit-tech.net/news/hardwar...rim-for-raid/1
Quick bench results:
http://img408.imageshack.us/img408/7802/benchu.png
It doesn't work like you think, TRIM in RAID that is - and probably explains why your 4K's appear to be so slow, slower than my single Intel 40GB that is.
Quote:
It will support TRIM with SSDs in an AHCI configuration, or with the RAID controller enabled and the SSD is used as a pass through device. An example of this use case is for users that want to use the SSD as a boot drive but still be able to RAID multiple HDDs together to allow for large protect data storage – a great use for the home theater PC. TRIM support for SSDs in a RAID configuration is under investigation and is not included in Intel® RST 9.6.
That's what I was going to post, but if you look in the comments section you'll see that people have debunked it
Quote:
It will support TRIM with SSDs in an AHCI configuration, or with the RAID controller enabled and the SSD is used as a pass through device. An example of this use case is for users that want to use the SSD as a boot drive but still be able to RAID multiple HDDs together to allow for large protect data storage – a great use for the home theater PC. TRIM support for SSDs in a RAID configuration is under investigation and is not included in Intel® RST 9.6.
so, in short, as Intel drives are pretty robust, and as I'm about to do a build (with Serious Sam) for his new rig using a budget of £160 for an 80gb SSD....
it would prolly be a good idea?
If I were to create a 60gb partition on the 2x 40gb RADI array, leaving the rest blank I think the TRIM issue goes away doesn't it?
True 4k, does is low, will have to investigate.
buy the 80gb and be done with, the 40 are limited and they will trim on a raid array so long as they are not in raid.
so you would lose the ability to trim them causing poor speeds in the future.
interesting potential though, no?
Some of the benchmark figures are off the chart ....
looks quite messy... will revert to the single 80gb... if it were a total no brainer I'd do it.. but have enough to be considering already....
this might be useful for some:
http://www.anandtech.com/show/3618/i...25m-g2-for-250
a really good review of 2x 40GB in raid0. note that in some cases it beats the mighty c300 :)
for the missing trim, I'm waiting what the future brings as I'm expecting that they will enable the trim function soon...
in my opinion it has a huuuge potential
btw, is there anyone who tried to do an image of the raid0 SSDs then manually erase the array and then restore the image? it would be kind of a hassle but afaik it would be necessary like once a month and it would resolve the performance degradation problem.