Hitachi's 500GB SATA-II Reviewed

Doggie Fizzle writes "The specifications for the Hitachi Desktstar 7K500 are impressive. 500 GB of disk space, 16 MB of cache memory, and 3.0 Gbps of transfer speeds are about as good as you are going to get in today's hard drives. The only category that might be rivaled is transfer speed, but that would require RAID or an Ultra320 SCSI drive to do so. This BigBruin review matches it up with some Seagate drives to show off its performance."

Comments

Do the differences matter for "most people"

[Mochatsubo]

For %95 of the population, do the specs of the latest and greatest matter?

Yes, yes, I know we are the 5%.

-m

Re:Do the differences matter for "most people"

[v1]

For those of us that run servers, rotation speed, seek time, transfer rate, these factors are really not that important. What counts in the trenches is how much space do you have, and how reliable is it?

Not everyone can afford a backup solution, some rely on raid protection, and others rely on a lucky rabbit's foot. Since I am in the 2nd category, (mirrors on anything that matters) I tend to actually look at cost per gb as the primary factor. If a drive fails, I send it in and get another one and resync the mirror. Every drive I buy has at least a 3 yr (if not 5) warranty. In the end, buying cheap drives is more cost effective than buying good drives, and is a lot more cost-effective than buying say a nice DLT drive and a pile of carts. (tho yes, mirror has pretty poor return on cost because of 50% usable space)

As long as I don't have to like swap out a drive more than once a year, I'm quite happy with reliability of even Maxtors. (though I still am not confident enough in my raids to install WD)

That being said, I wouldn't mind accquiring a pair of those 500's, though lately it's been getting a little tricky to find a FW bridge board that supports the really large drives. The last 300 pair I installed, (seagate even!) only one of the 14 bridge boards here would detect at 300. (instead of 128) Yes, they're all ATA6 and have up-to-date firmware, that doesn't seem to matter. WD uses their own "unique variation" on ATA6 for their big drives, so those are really fun to work with, I avoid them like plague.

Re:RPM ?

[nitehawk214]

RPM isnt the only factor. Remember that this 500GB drive has much higher data density on the platters. This means that it runs over more data in 1 revolution then a 100 GB drive.

Re:500GB finally?

[doormat]

Demand and the end of an underlying technology (GMR heads and parallel recording).

Not many consumers need 500GB of HD space in their computer for email and AOL. But 500GB would sure be useful in a Hi-Def PVR. But PVRs are still such a small segment compared to PCs.

Plus, tech wise, we're basically at the top of the S-Curve for the current HD technology. So we need to get the new technology and start the S-Curve all over again. We had a lot of advances when we went from 10GB HDs to 40 and 60GB HDs (one new larger capacity annoucement every quarter almost), but we've started to slow down and stagnate. I'm hoping things get going again soon and we make big advances from 1TB, 1.5TB. 2TB drives.

Re:LATENCY LATENCY LATENCY

[Gldm]

Unless the drives support an asynchronous write system, which they do. NCQ will reorder the writes anyway. Latency is primarly a read-side issue, random writes are not as common as random reads.

You can always improve your seek time by adding more redundant mirrors. If we apply the formula the formula seen here where x is the number of redundant mirrors, we can calculate the value of p which will give us our rotational latency for the mean seek time (hence the 0.5 because we want the 50% point for seek times).

Using this you can get 7200rpm drives to easily outseek a 15000rpm drive by using 4 or more redundant sources, and it's still cheaper for the same capacity, AND more failure tolerant.

This is why RAID always wins. Quantity has a quality all its own. SCSI used RAID to defeat the SLED concept in mainfraimes, commodity drives are doing the same to SCSI, by playing with the same rules.