Slashdot Mirror
Seagate Rolls Out 400 GB SATA Drives
SenorCitizen writes "Seagate is the first hdd manufacturer to announce 400 GB 3.5" hard drives. The 7200.8 is SATA native and comes with buffer sizes up to 16 MB. Seagate also announced a 2.5" portable external hard drive with 100 GB, and an external USB2 pocket hard drive with 5 GB. Get leeching!"
2.5" portable external hard drive with 100 MB
Wouldn't that 100GB?
"Bah!" - Dogbert
Seagate is not the first with 400GB disks,
IBM announced them a copule months ago and already ships them.
For large writes, it is, because the writes can be quickly cached and handed over to the drive for storage.
However, for most Linux hobbiests, small writes (saving conf files, storing UI tweaks, etc) are the name of the game. Wasting space with the cache like that is just asking for huge penalties for the normal Linux user.
Gimme a smaller cache and a cheaper price.
Get multiple drives and RAID them together. A 2-disc RAID-1 is quite reliable, but you can go for more if you are really concerned. Also, go SCSI instead of IDE. SCSI drives tend to be engineered to a higher standard, and are generally warenteed longer to boot.
However, don't bitch about the price. You WILL pay more for less storage, that's the cost of reliability.
Hitachi has had 400GB drives (SATA) for a few months now link
It looks like the only thing unique here is the "highest areal density", meaning (I assume) that Hitachi is using a four platter system, where Seagate's only has three.
Also, I wonder what problems might arise from 16MB caches on normal desktop machines. One of the issues I seem to recall with larger cache drives is the risk of filesystem corruption. If power is lost while data is sitting in cache, waiting for a write, then you could potentially royally screw up your file or filesystem. Hence, the only 16MB cache drives I've seen are notebook drives (almost always gonna have a battery) and SCSI drives (likely in a server or workstation, which will most likely have a UPS). Before you go countering that these aren't meant for desktop use, keep in mind that DV video, digital photgraphy, and music are all things that home users like the idea of, and they are also the things much more likely to consume massive amounts of storage capacity.
EB = Exabytes = BIGGG
I believe the problem you ran into is only during installs, and is similar to WinNT4's 4GB max boot partition. You can simply put the drive in another Win2K box that's already installed, format the full 160GB and use it nuts. Just be aware of NTFS versions that differ in Win2K/WinXP... I think XP has a newer version, and 2k can't use it, but could be wrong..
My Linux Command of the Day site : LCOD
That's the point of journaling, to ensure that the file system is never left in an inconsistent state, even in the event of a failure during write. You can loose data, of course, but that has always been true. Any data not comitted to disk will be lost, regardless of if it's in RAM or a disk buffer. However on a journaled file system (ext3, NTFS, etc) make it a near zero possibility that the file system be in an inconsistent state.
The default windows 2000 install does not support harddisk sizes over 128gb. SP3 enables the support for 48bit LBA, thus solving this problem.
Here's the related MSKB article.
The limit should be 2TB 2TB
EA David Gardner -"... but the consumers have proven that actually what they want is fun."
Link Has the info on how to get past that.
Higher density = more "bits - per - revolution"
More "bits/revolution" * same RPM = faster data rate
(of course if they just added platters, you wouldn't get faster - but it seems they're getting more bits by increasing the density/platter)
You are correct, and if you put an NT disk in an XP machine (say to do data recovery) the XP machine will -automatically and without asking- convert your NT disk to it's version of NTFS, rendering it unbootable.
-dameron
NewEgg.com (which is listed on pricewatch)
The length of the warranty shows how long they are willing to stand behind what they sell. If they will only stand behind it for 365.2425 days, then why should you trust it?
My IBM ThinkPad already allows me to boot from a USB Memory key.. and the ThinkPad is almost 2 years old.
What's the point? If the translator is good enough, you'd never notice a difference anyways. That's pretty much true of PCI-E and SATA, too.
-Erwos
Plausible conjecture should not be misrepresented as proof positive.
Just remember that RAID won't help you if you have a coupted filesystem, angry employee, or a virus. Tape or some other form of backup which can't always be accessed has advantages.
Most current BIOS's will already support booting off of USB drives.
It only applies to bulk hard drives in media processing devices (TiVos, DVD recorders, MP3 jukeboxes (set-top or portable), etc.), and it goes (basically) to the French **AA.
I found this out by RYourFA.
This tax was discussed by French's Ministry of Culture, but was vetoed by former Prime Minister Jospin. So, in fact, this tax never existed.
In my experience, Seagate's not been quieter (well, it's quieter than my old Bigfoot CY...) I've NEVER seen a Seagate fail, though, out of about 20 I've worked with. I've seen one Maxtor barf once, and the FAT was destroyed on the drive. The other five I've seen are fine. Western Digital is SHIT. I've seen (probably) five of them, and four have died. Hitachi/IBM? Seven of them, one was a 60GXP (shall I say more?), and the other fried the registry on Win2K when it was handled normally. As for Samsung, I've had decent luck with them - lots of damaged sectors on mine, though, but it ran Windows 98, and it was an HP preload (hey, this was before I knew Windows was crap, and same for HP).
6x400GiB in RAID5 would in fact be 2TiB. The general formula is (capacity) * (number of drives -1). For enterprise use this isn't necessarily true as you generally use larger numbers of spindles and leave one or more drives per set as hot spares.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
Actually 2^64 byes (1024 TB) equals a Petabyte NOT an Exabyte
Some cases will use these drives as internals...but they're a bit thicker than will fit in a standard 2.5" drive bay, so they have to put them into an external enclosure. Some older systems will fit bigger hard drives; in fact, I used to have a Pismo powerbook that I swear could fit TWO laptop drives in it. It was also twice as heavy as my TiBook.
Don't worry....regular 100 gig 2.5" drives are coming soon!
Hey freaks: now you're ju
It could take weeks.
Meanwhile, if another drive fails before the new one is built, then everything is lost.
you do realize that you could just take the floppy drive out of the system, its $10 addition to the system's price is nothing compaired to Microsoft tax you'd pay on the system anyway. Its not as if the floppy drive is holding back your system's performance, its integrated into the super I/O for goodness sakes and runs through the southbridge with all the other slow components.
Also most geeks i know have at least 100 floppies lying around that could be kept specially for emergencies while USB drives still cost too much to be kept only as a reserve for most people.
One important detail when constructing a multi-TB PC is that the 3Ware 8506 series cards can address at most 2TB per card. This is no problem with 250GB drives, but with 400GB drives, it becomes an issue (8 x 400GB = 3.2TB).
The recently announced 3Ware 9506 series can address more storage (4TB per card, I think) but when I looked two months ago, no vendors had it yet.
--Pat / zippy@cs.brandeis.edu
The number of writes that your average flash chip can endure is so large, that you'd have to TRY to "wear it out". Most flash devices these days (at least all the good ones) do wear patterning, where new writes are spread out over a large second of unused space (though this often requires filesystem support, see jffs[2]). This makes it even harder to kill flash memory since writes to one spot won't actually write to one spot, but will instead be spread over the whole device.
For grins, someone once went and calculated how long it would take to "wear out" the flash in a 32MB iPAQ. They figured out that, using wear patterning on an unloaded device (in other words, writing the the entire device and then starting over at the beginnign again), it would take 12 YEARS to wear it out if one were to write as fast as the flash was rated to write. Of course, if you're not wear patterning, you could wear out a sector quickly, but that's what filesystems have bad sector support for (even FAT has it, though that's a holdover from the ancient days where hard drives came with bad sectors, and of course flash filesystems support the inevitable wearing out of sectors).
Hitachi and Maxtor already had 400 gig drives out a while before this article was published. Though the maxtor doesnt have comparable specs. the hitachi does.
"Algebraic expressions are what we sue when we have no clue what were talkign about"
No drive does 50W. The 74GB 10k Raptor uses less that 10W. The Maxtor 15k Atlas SCSI (150GB) comes in under 15W. Takes less that 5 minutes to look up, no need to pull out wild numbers out of your ass.
"Hot lesbian witches! It's fucking genius!"