SlackwareThis Forum is for the discussion of Slackware Linux.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
Since you didn't mention type of service, concerns (performance and/or reliability) or cost factors I'd like to ask if this is either Enterprise use or in any way "mission critical"? Although Serial SCSI rarely comes with the commonplace 5 year warranty that Parallel SCSI used to have, they are still built more ruggedly both mechanically and electrically as well as having choices of extreme ruggedness (usually slower spindle speeds) or extreme performance (10K - 15K RPM drives are common) and SATA drives are 100 times more likely to encounter bit errors. The cost per TB is higher but 3TB SAS drives can be had for under $300 US with substantially higher performance and longevity over SATA.
Sorry, I should have included that in the initial posting.
This is for a personal NFS for videos, pictures, documents... etc. I am also running a Plex service from the server. As much as I would love some high end enterprise drives, I'd like to keep my cost per drive to under $120.
Most everything I have read in the past few years (quotes from two such articles are below) show SSDs are either about the same or slightly more reliable than HDDs. Failures were common when they first came out, but they've really improved over the years, which is why many datacenters are replacing HDDs with SDDs.
But for "raw speed and ultra high performance", you're still limiting yourself to the seek time of the drive. SSDs will surpass the performance of VelociRaptors for less price. Unfortunately, while the VelociRaptors are nearing the pinnacle of consumer HDD performance, they are easily surpassed by many quality SSDs.
But suggesting VelociRaptors doesn't fit the OP's request. There was no request on the speed of the drives, just capacity and quality.
All good points, but considering the amount of read-writes systems can do, and the fact I compile stuff a lot, an SSD drive for me will get more wear and tear rather than an HDD. Seek times may be a crutch of spinning disks, but those disks require much less specialized configurations to use with just about any file system. The durability of HDDs is just there.
All good points, but considering the amount of read-writes systems can do, and the fact I compile stuff a lot, an SSD drive for me will get more wear and tear rather than an HDD. Seek times may be a crutch of spinning disks, but those disks require much less specialized configurations to use with just about any file system. The durability of HDDs is just there.
Have you done any semi-recent research on this? SSD drives are not as frail as they used to be, even though many people still have this stigma about them. Back in summer of 2013, techreport got six SSDs (brands cover Corsair, Intel, Kingston, and Samsung) with the intention to write an insane amount of data to them every day until they die. Intel's 335 series 240GB drive was rated for 20GB of writes PER DAY for THREE YEARS (that equates to 22TB of data transferred), so that's what they set as the first benchmark.
They passed it with no problem, and didn't see the first signs of any problems until almost 5x the Intel spec (at 100TB they started seeing reallocation for one drive, but even then the drivewas functioning as it should and using the reserve NAND, as it's designed to do), and the first failures didn't occur until 700TB, which is 30x over the the spec! If we bump up the per day writing to 50GB/day, then that is over 5 years of constantly writing 50GB every day to the drive before you start seeing failures. One reached to over 2.4PB, which would would be 131 years if you stuck with 50GB/day!
Keep in mind, these are drives that are at least 2 1/2 years old. Newer drives will likely last even longer. I have my /, /home, /tmp, and swap all on my SSD, and I've had no hint of problems yet, but the speed was certainly a huge difference from my WD Black drive. (If I had more RAM, I'd probably move /tmp to RAM to speed up my compiles even more, but, alas, I am just stuck with 8GB due to a motherboard limitation.)
And they really don't require specialized configurations, considering both their write capabilities and support in various filesystems.
If you do go for the WD Green (I have 3 of the 3TB model and 2 of the 2TB model) - get the "wdidle3" utility and use it to disable the head unloading feature.
This will save a lot of head wear at the expense of some additional power consumption.
The 3 WD Green 3TB disks I own all had lost/bad sectors which led to media errors on reads within the first 3 weeks. Since two of the disks were mirrored, I was able to re-write the lost sectors from the mirror copy which fixed the media errors and avoided any data loss. The third disk had temporary data and I was able to overwrite the bad sectors with zeros which again fixed the media errors.
Bear in mind these disks have 4k physical sectors, so you need to write 4k (8 logical 512-byte sectors) to fix a media error.
After that they have all been stable, in approx 2.5 years of daily use (7997 power on hours = nearly a year).
This may not sound like an endorsement, but given the problems I've seen with other large capacity drives I don't think this is too bad given the price of the Green drives.
I would certainly recommend some sort of redundancy with modern high capacity drives, and regular backups too!
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.