Thread: what drive capacity would suit you?

  1. #76
    In my head happyclown's Avatar
    Join Date
    Dec 2008
    Location
    In my head
    Posts
    391
    Quote Originally Posted by Mario F. View Post

    It's just that I still fail to see any advantage. Show me how this could be advantageous.
    Well, if a single unit can do what a raid setup(several drives + cables + controller + available internal space inside the computer + required raid knowledge + setup time), then what do you think the advantage is?

    Quote Originally Posted by Mario F. View Post
    And think for a moment why it hasn't been done already. Surely we are well beyond our capability to build such drives. So... where are they? Or why haven't they became mainstream?
    I am sure the same questions must have been asked of Thomas Edison, Marconi, or Henry Ford before they came out with their inventions. Maybe your questions are ahead of it's time?

    Quote Originally Posted by Mario F. View Post
    How for instance you propose to solve the replacement of such a drive if one of its internal disks fails? Buy a new one? But just one of the internal disks failed...
    Notice I used "internal storage media", which means it doesn't have to be discs(platters). My line of thinking was, for simplicity, it could be something as hot swappable(pluggable) as a USB stick or an SD card. So it's not the drive that get's swapped, it's the internal storage media that get's swapped, like swapping the SD card in a camera.

    I will end my contribution to this particular topic with some of my favourite quotes from "Think & Grow Rich".

    "Those who are afraid of new ideas are doomed before they start".

    "Without doubt, the most common weakness of all human beings is the habit of leaving their minds open to the negative influence of other people."

    "MAN'S ONLY LIMITATION, within reason, LIES IN HIS DEVELOPMENT AND USE OF HIS IMAGINATION."

    I guess if someone can imagine a single unit drive with built-in redundancy, and they have the resources and capital to make it into a reality, they'd be called a pioneer. Until then, they're just plain KRAZY!!

  2. #77
    Master Apprentice phantomotap's Avatar
    Join Date
    Jan 2008
    Posts
    5,108
    I hate to tell you guys this, but you can already buy a "RAID-in-ONE" device and have been able to for years. You can get several different flavors. The reason you haven't heard about them often is the price point. You can spend $600 and get 96GB RAID5 (3.5 SAS FF). But why would you?

    Soma

  3. #78
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    Well, if a single unit can do what a raid setup(several drives + cables + controller + available internal space inside the computer + required raid knowledge + setup time), then what do you think the advantage is?
    To sum it up, the only advantage is that it saves the hassle.
    And the disadvantages?
    Less flexibility (you can't change the RAID configuration, add a disk into an array, or add spares)
    Higher possibility of failure (if the shared electronics fail)
    Cannot easily replace a disk

    People who have important enough data to secure will likely don't mind the hassle, and will want the higher flexibility and security.

  4. #79
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    Quote Originally Posted by happyclown View Post
    "Those who are afraid of new ideas are doomed before they start".

    "Without doubt, the most common weakness of all human beings is the habit of leaving their minds open to the negative influence of other people."

    "MAN'S ONLY LIMITATION, within reason, LIES IN HIS DEVELOPMENT AND USE OF HIS IMAGINATION."
    Nice quotes... which are also out of context.

    A curious trend I've been observing for some years now is that of "the incomplete perception of development" (if such a term is suiting). We are constantly offered products or solutions that advertise themselves as being radical, unique, the future, an improvement, development. It's marketing buzzwords, but that end up falling in the common language.

    And yet, we quickly come to realize there's nothing really revolutionary about many of them. It's like the Web 2.0 buzz. After all, what the hype likes is making sure it makes itself noticed. But Web 2.0 is nothing we didn't have for the past 20 years. There's really nothing new. Just a polished new look. So Web 2.0 could be instead a new leap in the web infrastructure, or the arrival of an HTML replacement, or even a whole new set of protocols offering new security levels to replace the old, crude ones like http, ftp., etc.

    Because of this, because we get into the habit of being happy with less, I feel that sometimes (many times) we just fall short on ambition. Something like a integrated disk array soon looks like the next best thing after the invention of hamburgers and anyone trying to point out there's nothing really revolutionary about the idea (and that the idea in fact is of little use) gets labeled as "being afraid of new ideas"

    I just don't buy it. And instead I throw your quotes back at you. I just happen to prefer dreaming of things that are really new and better, not rebranded and repackaged old ideas to fool the senses. I'm sorry but you just want too little. You are the ones fearing new ideas.

    EDIT: or.. to be crude, I'm in a whole other league, my friend.
    I have nothing further to add to this discussion. Except perhaps pointing you to phantomotap post above and offer you your own small uninteresting future. It exists after all. And as I predicted, much more expensive and not very useful. Enjoy.
    Last edited by Mario F.; 08-22-2009 at 07:35 AM.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  5. #80
    spurious conceit MK27's Avatar
    Join Date
    Jul 2008
    Location
    segmentation fault
    Posts
    8,300
    Quote Originally Posted by Mario F. View Post
    I have nothing further to add to this discussion. Except perhaps pointing you to phantomotap post above and offer you your own small uninteresting future. It exists after all. And as I predicted, much more expensive and not very useful. Enjoy.
    LOL as always.

    Just a note that you are distorting things quite a bit here, I presume just for the purpose of argument: my point was specifically about SSD drives which would make this feasible and cost effective, since it obviously would not be otherwise, as phantomap and others redundantly observed.

    I'm not sure why you would be so determined to "talk past" (ie, ignore) the point repeatedly but that is your SOP I guess
    Last edited by MK27; 08-22-2009 at 09:34 AM.
    C programming resources:
    GNU C Function and Macro Index -- glibc reference manual
    The C Book -- nice online learner guide
    Current ISO draft standard
    CCAN -- new CPAN like open source library repository
    3 (different) GNU debugger tutorials: #1 -- #2 -- #3
    cpwiki -- our wiki on sourceforge

  6. #81
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    As was answering directly to happyclown. Not you.
    You didn't try to make me feel like an unimaginative old soggy bag.

    But as for SSD, it is left to be seen if it would become cheaper. I suspect not. And there's always the issue of losing the flexibility in maintaining and upgrading your array without any real benefit to counter that.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

  7. #82
    spurious conceit MK27's Avatar
    Join Date
    Jul 2008
    Location
    segmentation fault
    Posts
    8,300
    Quote Originally Posted by Mario F. View Post
    But as for SSD, it is left to be seen if it would become cheaper. I suspect not. And there's always the issue of losing the flexibility in maintaining and upgrading your array without any real benefit to counter that.
    I know next to nothing about the technology, I was just guessing that it would make this much more feasible than with conventional drives.

    I'm sure it wouldn't be a "high end" option either. If you are a computer professional, of course you might as well just set up your own array. But if you owned a small business and could buy a "2 in 1 back-up drive" cheaply I'm sure people would do it.
    C programming resources:
    GNU C Function and Macro Index -- glibc reference manual
    The C Book -- nice online learner guide
    Current ISO draft standard
    CCAN -- new CPAN like open source library repository
    3 (different) GNU debugger tutorials: #1 -- #2 -- #3
    cpwiki -- our wiki on sourceforge

  8. #83
    and the hat of sweating
    Join Date
    Aug 2007
    Location
    Toronto, ON
    Posts
    3,545
    Here you go: Canada Computers - Your Neighbourhood Computer Store and Service Centre - PC Systems and Hardware Components, Notebooks, Electronics, and more.
    You can put 2 or more drives in a box with RAID built in for about $60+.

    As for SSD's, I could see how they might design them to do optional RAID by partitioning the memory chips, and if a chip fails, it could automatically hot swap over to another chip... but it still has a single point of failure (i.e. the main circuitry that controls all the chips). Maybe some less intelligent home users might see all the marketing for such a device and say "yay, now I won't need to worry about doing backups anymore..." but IT professionals should know better and use normal RAID that removes the single point of failure.
    "I am probably the laziest programmer on the planet, a fact with which anyone who has ever seen my code will agree." - esbo, 11/15/2008

    "the internet is a scary place to be thats why i dont use it much." - billet, 03/17/2010

  9. #84
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    SSDs' internal "RAID" doesn't give redundancy. They are for speed (RAID-0).

  10. #85
    Registered User
    Join Date
    Apr 2004
    Posts
    210
    Quote Originally Posted by cyberfish View Post
    To sum it up, the only advantage is that it saves the hassle.
    Which is a major point for the most of its potential users - which would not be sysadmins but homeusers who wouldn't want to lose their data. I.e. my dad with his 50GB of digitized holiday/vacation/birthday/wedding/... photos (which took him half a year to scan). He's backing them up manually now and doesn't have the first clue about "RAID-what(!?)".
    I know he'd get one of those "raid-ssd"s if he could simply plug them in like a normal HDD and never think about it again. Twice the price, no matter.

    Quote Originally Posted by cyberfish View Post
    And the disadvantages?
    Less flexibility (you can't change the RAID configuration, add a disk into an array, or add spares)
    Absolutely inconsequential for most users. I don't want to throw out numbers here, so I'll go with "most" and I think we can agree on this: PC's are pretty much in every home today. They are a tool for most people now, used for home-shopping, email, printing and the occasional game. Bought in the shopping mall by people who'd ask you "raid-what? where?". Everyone has data that he wouldn't want to lose.

    Quote Originally Posted by cyberfish View Post
    Higher possibility of failure (if the shared electronics fail)
    Cannot easily replace a disk
    This would obviously depend on the implementation. If the media were replaceable, neither would be true.

    Quote Originally Posted by cyberfish View Post
    People who have important enough data to secure will likely don't mind the hassle, and will want the higher flexibility and security.
    No. Don't project your curious, knowledge-seeking programmer's mind onto everybody else. You're making assumptions based on your personal interest, but most people really couln't care less how their computer works - as long as it does.
    main() { int O[!0<<~-!0]; (!0<<!0)[O]+= ~0 +~(!0|!0<<!0); printf("a function calling "); }

  11. #86
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    Quote Originally Posted by Nyda View Post
    Which is a major point for the most of its potential users - which would not be sysadmins but homeusers who wouldn't want to lose their data. I.e. my dad with his 50GB of digitized holiday/vacation/birthday/wedding/... photos (which took him half a year to scan). He's backing them up manually now and doesn't have the first clue about "RAID-what(!?)".
    I know he'd get one of those "raid-ssd"s if he could simply plug them in like a normal HDD and never think about it again. Twice the price, no matter.

    Absolutely inconsequential for most users. I don't want to throw out numbers here, so I'll go with "most" and I think we can agree on this: PC's are pretty much in every home today. They are a tool for most people now, used for home-shopping, email, printing and the occasional game. Bought in the shopping mall by people who'd ask you "raid-what? where?". Everyone has data that he wouldn't want to lose.

    This would obviously depend on the implementation. If the media were replaceable, neither would be true.

    No. Don't project your curious, knowledge-seeking programmer's mind onto everybody else. You're making assumptions based on your personal interest, but most people really couln't care less how their computer works - as long as it does.
    We agree that it will only make sense if the media is replaceable? By replaceable, do you mean user-replaceable or manufacturer-replaceable?

    If the media is not replaceable, both flexibility and security will be a lot worse.

    If it's user-replaceable, we are talking about something very different from modern harddrives, now. If we can replace the media, why can't we just replace the whole disk? How about using a RAID enclosure and actual harddrives as the "media"? They are "zero-configuration", too.

  12. #87
    Registered User
    Join Date
    Apr 2004
    Posts
    210
    Quote Originally Posted by cyberfish View Post
    We agree that it will only make sense if the media is replaceable? By replaceable, do you mean user-replaceable or manufacturer-replaceable?
    No, I meant we agree that there is a huge market of home users with little to no knowledge about computers or how to backup their data properly.

    Do you think media replaceability would be important? HDD capacities double every year or so, right? (Just a guess, I don't have the numbers.) I think most home users wouldn't mind getting a completely new, larger drive at the occasion and be happy that they got a "The system drive is failing. Please backup all data and replace the drive."-message instead of total data loss
    If the drive fails in less than 2 years, they'd get a free replacement anyways, so they really couldn't care less if only the manufacturer can replace the media.
    main() { int O[!0<<~-!0]; (!0<<!0)[O]+= ~0 +~(!0|!0<<!0); printf("a function calling "); }

  13. #88
    Registered User
    Join Date
    Dec 2006
    Location
    Canada
    Posts
    3,229
    Quote Originally Posted by Nyda
    Quote Originally Posted by cyberfish
    Higher possibility of failure (if the shared electronics fail)
    Cannot easily replace a disk
    This would obviously depend on the implementation. If the media were replaceable, neither would be true.
    If they were not replaceable, both would be true .

    Do you think media replaceability would be important? HDD capacities double every year or so, right? (Just a guess, I don't have the numbers.) I think most home users wouldn't mind getting a completely new, larger drive at the occasion and be happy that they got a "The system drive is failing. Please backup all data and replace the drive."-message instead of total data loss
    And what advantage does it have over RAID "zero-configuration" enclosures? Probably not any cheaper, will be less flexible (cannot upgrade or replace the disks), and less secure (if the shared electronics fail). I think "Please buy 2 identical SATA harddrives of your desired capacity and put them in the slots" (at least they can just show the box to a computer store sales, who will show them their options) is user-friendly enough? Or are you saying people are too lazy to even do that, and will be willing to accept the aforementioned disadvantages in a tradeoff?

    People can even sell enclosure + 2 drives bundles.

  14. #89
    Registered User
    Join Date
    Apr 2004
    Posts
    210
    Quote Originally Posted by cyberfish View Post
    Or are you saying people are too lazy to even do that, and will be willing to accept the aforementioned disadvantages in a tradeoff?
    Absolutely, yes. But for the sake of the argument, just say the media would be replaceable (no hotplugging - more like RAM is replaceable).
    I don't really see the value in this, though. If a media fails, you can still access your data and copy it to an entirely new drive. Given at least one year of guarantee (not sure how realistic it is to make a successfull warranty claim after 2 years) it's probably more efficient to get a completely new and probably a lot faster and larger drive. Especially since the advancements in ssd technology are rather fastpaced at the moment.

    Lack of flexibility isn't a disadvantage if all you want is a safe drive. It's a big advantage since it also means it's easier to set up.
    I'm not entirely sold on the reduced security either since I doubt there is a lot of electronics to share on an ssd. The physical casing would be shared of course, but that's not an issue unless you're into crash-testing stuff with your harddrives.

    Anyway, that's just my opinion. Whether or not manufacturers will actually ever produce these devices is an entirely different matter. Happyclown and mk27 seem to think so. I'm not sure
    main() { int O[!0<<~-!0]; (!0<<!0)[O]+= ~0 +~(!0|!0<<!0); printf("a function calling "); }

  15. #90
    (?<!re)tired Mario F.'s Avatar
    Join Date
    May 2006
    Location
    Ireland
    Posts
    8,446
    I'm not sure I'm following the debate.
    RAID 1 implements just the needed functionality. If a drive fails, the other takes control and the array is disabled with nothing more than a notice to the user letting them know of the failure. Since RAID 1 can be implemented with zero-configuration, how come this is problematic to non savvy users?

    And they have the added benefit of buying a new hard drive and have someone plug it for them at their local pc repair shop at a minimum cost. Losing an integrated array and replacing one of its internal disks is bound to be more complex and expensive.
    Last edited by Mario F.; 08-23-2009 at 10:48 AM.
    Originally Posted by brewbuck:
    Reimplementing a large system in another language to get a 25% performance boost is nonsense. It would be cheaper to just get a computer which is 25% faster.

Popular pages Recent additions subscribe to a feed

Similar Threads

  1. SATA HDD failure
    By PING in forum Tech Board
    Replies: 4
    Last Post: 12-23-2008, 12:25 AM
  2. Bitwise Unwanted Output
    By pobri19 in forum C++ Programming
    Replies: 4
    Last Post: 09-15-2008, 04:07 AM
  3. Error with a vector
    By Tropicalia in forum C++ Programming
    Replies: 20
    Last Post: 09-28-2006, 07:45 PM
  4. finding capacity of hard drive
    By jverkoey in forum Windows Programming
    Replies: 4
    Last Post: 08-20-2003, 07:21 AM
  5. Retrieve total drive capacity
    By Markallen85 in forum Windows Programming
    Replies: 3
    Last Post: 08-18-2003, 12:44 PM