Forgot your password?
typodupeerror
Data Storage Networking Software Linux

Distributed Storage Systems for Linux? 52

Posted by Cliff
from the spread-out-all-ova-tha-place dept.
elambrecht asks: "We've got a _lot_ of data we'd like to archive and make sure it is accessible via the web 24/7. We've been using a NetApp for this, but that solution is just waaaay to expensive to scale. We want to move to using a cluster of Linux boxes that redundantly store and serve up the data. What are the best packages out there for this? GFS? MogileFS?"
This discussion has been archived. No new comments can be posted.

Distributed Storage Systems for Linux?

Comments Filter:
  • Lustre (Score:5, Informative)

    by Yerase (636636) <randall.hand@nOsPaM.gmail.com> on Thursday May 05, 2005 @12:15PM (#12442208) Homepage
    Check out Lustre at http://www.lustre.org/ [lustre.org] It's being developed/used by the DOE on alot of Supercomputer Cluster systems, for multi-terabyte storage stuff.
    • It looks great, except for one thing:

      2.4 only.

      The system itself is developed in a VERY cathedral-like style by a company called ClusterFS, who is selling the 2.6 version. My guess is they'll release it for free when Linux 2.8 or 3.0 is released, so that they always give away the obsolete version and sell the new one.
      • What, pray tell, does 2.6 offer that makes it needful for the questioner's application?
        • It's just a lot nicer in a lot of ways. What, pray tell, does Lustre offer that makes it "needful" in that situation? He could just get some sort of a SAN device. But Lustre would probably be nicer. In the same way, 2.6 is nicer.

          I've seen systems still running 2.2, but it's not pretty. 2.6 is the way to go these days. Lustre is about the ONLY reason anyone should be using anything older.
  • It would be useful to know about how much data we're talking.
    I suppose there's a difference between serving just 500GB or a few terabytes.
    • Also, what else do you need to do with the data? Back it up? Mirror it? Data mine it? Any type of performance requirements? Interoperability requirements? The list goes on. We need more information.
    • > I suppose there's a difference between serving just 500GB or a few terabytes.

      If it is only that, you could just stick 8 disks in a nice server system, that gets you a few terabytes. Use your filesystem of least mistrust, such as reiserfs, XFS or JFS.

      RAID 1+0 or 10 is very nice, but you still have a single point of failure. NetRAID and some kind of fail-over redundancy with another machine way be the way. You will not get a full SAN/NAS this way, but you also avoid a lot of complexity.
  • by middlemen (765373) on Thursday May 05, 2005 @12:35PM (#12442472) Homepage
    Panasas http://www.panasas.com/products_overview.html [panasas.com] has some products which probably fit your requirement of high speed distributed storage.
    • Panasas http://www.panasas.com/products_overview.html [panasas.com] has some products which probably fit your requirement of high speed distributed storage.

      Panasas is a GPL Violator, as per conversations with them during a demo. They have a proprietary module that is built against only RHEL kernels. They confirmed that this module uses kernel headers and other parts of the kernel, but they refuse to release the source to any customer.

      If you are ever in a situation where you need to upgrade kernels in a hurry, switch
  • by afabbro (33948) on Thursday May 05, 2005 @12:38PM (#12442500) Homepage
    What kind of idiotic Ask Slashdot is this? All of the important data is missing:
    • What's "a lot"? 1MB is a lot of data if you think about it. When people start talking about "a lot" of data these days, I assume they're meaning hundreds of terabytes. Is that what you mean?
    • What's the budget? What performance do you need? Do you need to back it up? Do you need to replicate it? Your post is sort of like "hi, I have a problem. What is the answer? Thanks!"
    Also, it's "too expensive to scale," my friend. You'd think an "Editor" like Cliffy would fix posts, but he's too lazy.

    If you can afford NetApp, why not keep with NetApp? A bunch of Linux boxes is not a storage solution. Indeed, what does Linux have to do with anything? We're talking storage here. What are you planning to do - put in 200 of them with internal SATA drives? Yeah, that'll be a lot cheaper to maintain...

    I'm not shilling for NetApp, but if you really have "a lot" of data to put "on the web" "24/7" then you need some kind of real storage solution like a NetApp or one of their competitors.

    Now go away and please take Cliff with you.

    • NetApp is number four in storage revenue terms, after EMC, HP and IBM

      so go ask them about what you want

      really you can admin your white box's (that become a NAS ) or you can get a NAS

      are you thinking SAN ?

      also talk to Apple they do some nice product as well as SUN

      whats this for large data ?
      video data go talk to SGI and their XFS products

      really it depends on what your doing NetApp is great for company File system of documents but Bad if you want to get the most out of your storeage and you do mostly vid
      • SAN is the ultimate storage solution. IBM, HDS, EMC, HP, SGI, Engenio go with any of them. Go with the cheap lineups if you have to.

        NAS/NetApp is so overrated, more like 7th place. Only reason why it has made a headway is because it is so close to NFS. And everyone can do NFS commands.

    • A bunch of Linux boxes is not a storage solution.

      Hey man, don't tell that to Google.
    • Also, it's "toO expensive to scale," my friend. You'd think an "Editor" like Cliffy would fix posts, but he's too lazy.

      I don't think the Slashdot editors can spell either. An "editor like Cliffy" recently allowed the word "persue" to appear in a headline.

  • Back about 4 years ago we were forced to get a maxtor netattach (can't remember the name) because at the time journaling file systems were virtually non-existent. Then that lasted for a year before we outgrew it and then we went with a Dell NAS server 600GB also windows 2000 embedded. It has scsi connection to connect SuperDLT tape backup drive and the windows 2000 backup program works for our needs. Simple for any average joe to restore files.

    I did look into getting a linux NAS but the solutions out t
    • Snap Appliances would seem to meet your requirements.

      They are linux based, web management, have the scsi port, and they sell backup and restore software.

      I use one, but I back it up as a linux server with Veritas Netbackup to an existing tape robot on another linux server.
  • AFS ?? (Score:4, Interesting)

    by forsetti (158019) on Thursday May 05, 2005 @12:52PM (#12442660)
    How about OpenAFS [openafs.org] ? It is sort of like NFS on steroids, with redundancy, scaling, cacheing, Kerberos-based security ... I've just started looking at it myself, but it seems pretty slick.
    • AFS is a distributed filesystem with highly integrated clustering. What catches many people off guard about AFS is the way it does replication. Volume replicas are read-only, and in order to update the replicas, a command must be issued (vos release) - they are not automatically kept up-to-date. As long as your replication needs are read-only and you don't need coherence automatically maintained for you, AFS might be the way to go.
      • Lacking read/write replica support can be annoying at times, but the manual release of the read/write copy to the replicated ro clones can be damn handy. Think about it - you get a free builtin staging area for whatever data you are pushing out! I use it that way, at least, for staging new versions of websites and applications before "releasing" them to the replicated read-only sites that are visible to the world.
  • Centera (Score:5, Informative)

    by egarland (120202) on Thursday May 05, 2005 @01:11PM (#12442927)
    Get A Centera.

    I'm biased but this is a high level Linux based storage system done right. It's not easy to create a coherent storage system out of lots of separate machines, the software that runs on this cluster does a lot of work. This thing fully redundant with no single point of failure, dynamically expandable without even taking it offline, it scales to 100's of terabytes and manages all that content continuously (scanning for corruption and fixing it, garbage collecting, etc..). The cluster has redundant backend networks and parallel paths everywhere, it even uses reiserfs to store the data. There's a lot of good engineering in this unit and they sell it at a decent price compared to NAS boxes.

    Check it out:
    http://www.emc.com/products/systems/centera.jsp [emc.com]
    I do work for EMC (like I said.. I'm biased) but I don't speak for them, my opinions are my own.

    Storage clustering is simply hard to do while still presenting a low level filesystem interface. Tossing that out and creating file storage as a high level service with a richer interface seems like the right approach to me. Show me a storage clustering solution that doesn't do that and I'll show you something full of bugs, expandability issues, limitations, and pain points.
    • Not to knock your employer (ok, so maybe to knock your employer). But the Celerra has soured my tast for EMC NAS (or NAS type) solutions so much that even without ever working with NetApp, I would recomend them.

      Of course, the centera always did sound interesting. But last I heard you still needed to write to the centera API, no block access for you (or real NAS type either). But I did hear murmers of this changing.

      Anything to avoid a Celerra.
      • But last I heard you still needed to write to the Centera API, no block access for you (or real NAS type either). But I did hear murmers of this changing.

        There are at least two products you can put in front of Centera to make it look like a standard filesystem: CUA (a Centera specific EMC product) and Legato Disk Extender. The tradeoff is that by interfacing with it like a filesystem you re-introduce the limitations of filesystems and lose all the automatic functionality the API gives you. If you look a
    • As long as we're being honest...

      I was just laid off from Xiotech,(with about 100 others) one of the smaller SAN vendors..

      But we call the Centerra a "data jail". It's like the roach motel..

      Data checks in, but it don't check out. It can't scale beyond a 42U rack enclosure. It's a bunch of little servers striped together to form a big NAS with a metedata controller in the middle.

      Just bashing the 800lb gorilla...
      OTOH, If you're hiring, I'm willing to tell you your products could rule the world!

      Regards.
      P
      • Re:Centera (Score:3, Insightful)

        by egarland (120202)
        But we call the Centerra a "data jail". It's like the roach motel..

        Ug. It's just not true. Most applications that are built to work with Centera include functionality to migrate in/out of the system just like most applications that are built to work with tape can both put data on and get it back. The difference is tape sucks, Centera doesn't.

        It can't scale beyond a 42U rack enclosure.

        Also not true. I have worked extensively with a 3 rack install with about 50tb of data on it. I believe all versi
        • Oh yea.. I forgot the obligatory..

          I work for EMC but I don't speak for them, my thoughts are my own even if I sound like an EMC cheerleader/sock puppet.

  • Ask Google (Score:3, Interesting)

    by VernonNemitz (581327) on Thursday May 05, 2005 @01:24PM (#12443088) Journal
    I'm sure they'd be happy to sell you something along the line of serving data....
  • I did some reasearch on clustering filesystems for work a while ago. Here's the Cliffs-notes version:

    GFS
    High-end, a pain in the ass to set up and run. Wants a RHEL server or two to run.
    OpenGFS
    Started as a fork of GFS when the GFS license changed, it has followed a bit of a different path. Not nearly as stable or fast as GFS, but might be there some day.
    Lustre
    Lustre should be really nice, but is horrendous to run (at least, that's the word from my friends at Sandia, who know a thing or two about it). General consensus is that you need a full-time staff member just to make it work. If you can afford that, it's a good way to go.
    PVFS
    Fast, light-weight, not POSIX-compatible. If your apps don't need the stuff it doesn't do, or you're willing to write some glue code for your app to speak PVFS natively instead of using the FS driver, this is a great way to go. Looks simple to set up (as simple as these things get).
  • by Dr.Dubious DDQ (11968) on Thursday May 05, 2005 @01:49PM (#12443445) Homepage

    A barely-related subject - I've been wondering whether there's some way to collect the unused space on all the Windows workstations around here into a shared space for storage.

    This is purely a speculative exercise, but I keep wondering if some combination of:

    • Every Windows(tm) workstation "shares" an otherwise-empty subdirectory
    • a Linux box creates and uses a "filesystem image" file of some kind ("loopback mount"-style image) stored on each share over SMB/CIFS
    • Linux uses VFS to combine the individual virtual drives into a larger drive (or perhaps two identical-size virtual drives, which are then combined into a single software RAID 1 array?)
    • Linux then shares this Rube-Goldbergian system as a Samba share...

    Yes, I know it's kind of silly, and performance seems like it would be pretty pathetic, but the more I think about it, the more I want to see if I could actually do it (think pretty much the same mindset that the IP-over-carrier-pigeon guys had...)

    Heck, it might conceivably actually WORK for a large-but-infrequently-accessed historical repository or something...

    Or has someone already started some sort of "Virtual ATA-over-ethernet-from-a-file driver for Windows" project and spoiled my fun?...

    • If you want to try building it, I'd suggest you start with a nice high level method of creating linux based filesytems:

      http://perlfs.sourceforge.net/ [sourceforge.net]

      Build it first, optimize later.

      FYI.. The multi-threaded filesystem version exists, I just haven't bundled it up pretty for distribution. Now someone needs to create a multi-threaded samba to share it out.
      • That's actually the part I'm not sure about - I know I could e.g. format an old 6GB HDD and then use dd to make a filesystem image that I could mount, but I haven't done any digging to find out if it's possible to directly create a ('standard') filesystem as an image file. (Hints welcome...)

        Perlfs looks interesting but it appears as though it hasn't been updated in a while (the homepage talks about adding support for linux "2.5" at some point...)

        • [...], but I haven't done any digging to find out if it's possible to directly create a ('standard') filesystem as an image file. (Hints welcome...)

          Huh? Just run mkfs.whatever on your file. Should work without problems. Your filesystem is as large as it would be on an equally large blockdevice.

          Example:

          $ mkfs.ext3 file
          mke2fs 1.36 (05-Feb-2005)
          file is not a block special device.
          Proceed anyway? (y,n) y

          Filesystem label=
          OS type: Linux
          Block size=1024 (log=0)
          Fragment size=1024 (log=0)
          1784 inodes, 7116 blocks
        • Perlfs looks interesting but it appears as though it hasn't been updated in a while

          Sorry about that. I'm busy. You did get me motivated to put up the new multi-threaded capable version of perlfs today though. Maybe I'll even edit the home page at some point.

          And yes.. it's somewhat actively maintained. No, I haven't worked on it with the 2.6 kernels yet. :)
    • first off all, iSCSI not samba

      starwind will allow you to export images or even drives and partitions to iSCSI. WAY less overhead and way faster than smb. I can only get about 6MB/sec over 100 speed ethernet and SMB, but i can get ~10.5MB/sec with iSCSI.

      simply add a number of these disk images accross your network and then mount them on your linux system. you can use LVM to manage these volumes also.

      i have to warn your that 100 speed ethernet is a bit slow for data storage on a network especially when y
  • We use OpenAFS (Score:4, Insightful)

    by Bamfarooni (147312) on Thursday May 05, 2005 @01:49PM (#12443446)
    We have about 27TB of data from Mars (and adding another TB per month) that we need to keep online. We have been using netapps, but at ~$25K/TB, plus maintenance (3 years maintenance is about as much as a whole new system) they're just WAY too expensive for data warehousing.

    We've moved to using linux based OpenAFS servers. A high quality 3U box (qsol.com [qsol.com]) loaded with 16x 300GB ATA drives costs about $8.5K and provides us about 3.5TB (2 drives for parity, 2 drives for hot-swap). That works out to $2.5K/TB. If your risk tolerance is higher than mine, you can bring that up to $8K/5.5TB, for about $1.5K/TB). We really want 99.999% availability, so just to be safe, we keep a 100% redundent read-only copy on a second machine (AFS supports this beautifully, including automatic fail-over).

    OpenAFS has a couple of features that make it better than NFS (client-side cache, for instance), but it also has a few drawbacks, like no files >2GB.
    • Re:We use OpenAFS (Score:2, Informative)

      by luizd (716122)
      Not anymore in OpenAFS 1.3.81.

      Copied from release notes:

      For UNIX, 1.3.81 is the latest version in the 1.4 release cycle. Starting
      in 1.3.70, platforms with pthreads support provide a volserver which like
      the fileserver and butc backup system uses pthreads. Solaris versions 8
      and above, AIX, IRIX, OpenBSD, Darwin, MacOS and Linux clients support
      large (>2gb) files, and provided fileservers have this option enabled.
      HP-UX may also support large files, but has not yet been verified. We hope
      sites which can do s
    • heh. you must be Z, E, or C

      hello from M over in PSF =)
      • nevermind. found out you are neither of the three <G>

        i was talking to Z at the installfest 2 weekends ago and he told me about this migration. it sounded pretty cool. we have been eying a similar migration too (primarily for authentication, not storage), but we're waaaaaay short on resources here =(
  • If your data is partitionable into small-enough discrete units that have low or not inter-unit dependencies, then it should scale almost without limit.

    After all, as a collection there is an immense amount of data on "the world wide web" but since it's partitioned, scaling isn't an issue.

    Even before the web, the universe of ftp, gopher, news, and other servers held gobs and gobs of data, nicely partitioned.

    When answering questions like this, it would help to know the organization of the data and if it can
  • Check out http://www.ibrix.com/ [ibrix.com] This is a perfect solution for your requirements. Pixar uses this.
  • aRchive.org (Score:3, Informative)

    by fulldecent (598482) on Thursday May 05, 2005 @02:42PM (#12444052) Homepage
    This is the solution archive.org uses.

    http://www.archive.org/web/petabox.php [archive.org]

    They are on the order of petabytes
  • There's a great article on ATA over Ethernet (AoE) and it has a story about a guy who put 2TB of RAID 10 up for $6,500. It looks like a fascinating solution for storing large volumes of data. If your data is primarily static, a couple of these machines replicating between themselves and you're good to go.
  • "We've got a _lot_ of data we'd like to archive... We've been using a NetApp for this, but that solution is just waaaay to expensive to scale. "

    The theoretical upper limit of any file system is limited by 2 things, the address space, and the efficiency of the data structure.

    In a 32 bit system, that means that, in theory you could fit 4.2 billion objects into a file system... but don't try it. NTFS craps out at between 15 and 50 million depending on whose numbers you are willing to listen to, EXT3 sta

  • um, the stats speak for themselves ~ 64-bit scalability to support files to 9 million Tb, filesystems to 18 million Tb ~ Instant data sharing without network mounts or data copies among all major OS: IRIX® Sun(TM) Solaris(TM) IBM® AIX® Windows® (2000/XP) Linux® (32 and 64 bit) Mac OS® X Unix® Flavors # Highly optimized distributed buffering techniques that provide the industry's fastest performance # High availability with automati

Organic chemistry is the chemistry of carbon compounds. Biochemistry is the study of carbon compounds that crawl. -- Mike Adams

Working...