joe.to forums

i don't fully understand how this works





i can't tell if it's invisible to the OS or how it knows what the "most used files" are without having access to the OS. i don't really understand what the hardware does at all that couldn't be done with software

but it sounds interesting

I heard about this, supposed to be freaking awesome.

I can see what you are saying Joe, the way I can see it has it as cache.

Don't know to much about it.

I think the hard drive can know what files are the most commonly read, and auto-move them to the SSD. It probably stores the metadata on the SDD, and looks at the metadata to see if the data is on the SDD, on the HD, or on both.

With this you would be able to fill the full sata pipe from the SSD and copy data from the SSD to the hard drive...but then wouldn't you have (slightly) higher latency because it has to travel though another thing? Maybe for computers with the 1.5Gb/s SATA this is better because that pipe is easier to fill, but if you have the 3.0Gb/s or 6.0Gb/s I think it would be better real-world to have the 2 drive connected.

Price seems to be under $100 according to random thing I read on the internet.

It just copies the first X amount of your HDD over to the SSD where X is the size of the SSD, it will then read from the SSD any time that data is requested and sync that data at each reboot. There are ways to defragment a standard HDD with most used data at the front or the most important folders, which is what you would need to do since it is transparent to the OS.

Here is some performance numbers, and the silverstone website has the documentation up.

but what is the hardware doing? they could just do that with software more efficiently

it is transparent to windows so you can literally buy that put a SSD disk in and it would make certain parts of the disk faster, yes you could do a RAM driver or something along those lines but that takes time and know how that would just work.

i still dont get it

lets say you have two sata drives - a 1TB disk drive and a 100GB SSD

you connect 1TB to sata1
you connect this device to sata2 and then SSD to this device
your system says you have 1TB? not 1TB+100GB
then as you use the 1TB on sata connector 1, how does sata connector 2 see what sata connector 1 is doing? it doesn't interface with OS, in fact it would be 100% idle if OS doesn't see it.

the chip on the card does a RAID 1 mirror of the first 100GB of data from the 1TB drive and any time that data is requested it is read from the 100GB SSD giving lower access time and greater speed. The operating system Only sees 1TB of storage but not the identifier of the 1TB HDD, it changes to that of the HDDBOOST from what I have seen which is just a modified RAID 1 chip or a backplane with 1 SATA in 2 SATA out, except this one does a modified RAID.

The documentation says it does the RAID build on boot but there isn't any good documentation on weather or not it does a differential while the computer runs or not and no data on what happens as the data changes either.

so both the disk drive and the ssd are both plugged into this device? if not how can it assume control of the drive on the other sata controller?

yes both are connected to that device, the SSD is placed inside the tray and the normal HDD is connected via a SATA cable

update it's out now

http://www.legitreviews.com/article/1361/1/

apparently it COPIES the first nGB of your HDD to the SSD and filters all reads to the SSD and all writes get mirrored to both

according to the review this works very well. however i would have some reservations - what if my most frequently accessed files aren't in the front of the HDD? this would be almost useless.

Defrag the HDD with a custom rule to put your "needed" files at the front

yeah but that's not foolproof and plus it makes this thing more of a hassle than just using the SSD as a conventional additional drive

if you have MyDefrag on a schedule just make it so that it uses a custom profile seems like a good way to me