User:Tim/NFS server build

From Van Essen Lab

< User:Tim
Revision as of 16:30, 20 June 2011 by Tim (Talk | contribs)
Jump to: navigation, search

The goal of this project is to create 2 identical NFS servers from standard parts, having roughly 20-30TB of robust storage each, for a cost of under $4000 each. One of the two will be built first to ensure compatibility of parts, and to make any necessary adjustments to the build, following which the other will be built to the final working specifications of the first.



Storage: 10 or 12 of one of the following:

Western Digital Caviar Green 3TB - 5400rpm, $150, cheapest 3TB from Western Digital

Hitachi Deskstar 3TB - 7200rpm, $180, cheapest 7200rpm 3TB on newegg

Hitachi Deskstar 3TB - 5400rpm, $130, cheapest 3TB on newegg

Seagate Barracuda XT 3TB - 7200rpm, $215, cheapest Seagate 3TB on newegg

Boot drives: 0 or 2 of one of the following:

OCZ Vertex 30GB SSD - $85, cheap but reputable solid state drive with okay performance (per Tom's Hardware Guide) and enough size for OS, will outperform rotational disks for booting

OCZ Vertex 2 40GB SSD - $105, slightly more space and significantly higher specifications than the above OCZ Vertex

OCZ Vertex 2 60GB SSD - $120, slightly more space, and better sustained write, nearly equivalent to Vertex 2E series

OCZ Agility 3 60GB SSD - $135, SATA III boosts speed to nearly double the Vertex 2, if motherboard supports SATA III

Seagate Barracuda 250GB - $37, cheap (slow) hard drive to let the OS boot without the SATA controller expansion card operating


Lian Li aluminum ATX toolless full tower - $320, toolless case, recommended by Jon S

Lian Li aluminum ATX full tower - $290, similar case, but requires tools for assembly.

XCLIO steel chassis ATX full tower - $210, cheaper case with a front made of 12 5.25" bays, comes with internal racks for 4 hard drives in the space of 3 5.25" full height bays, conceivably could fit 16 hard drives, but without hotswap bays

Drive bays

10 or 12 of Kingwin hotswap rack - $19, has activity and power lights for each disk, activity light will help for locating a failed disk.

Power supply

PC Power and Cooling "Silencer mkII" 650W - $95, single high current 12V rail at up to 46 amps, 12 WD Caviar Green drives only need 21 amps if all 12 spin up at once

System configuration

Instead of a proprietary embedded OS with a web interface, this machine will run a full-fledged desktop operating system, likely ubuntu 10.10, for ease of maintenance. Hardware raid controllers are expensive, and tend to use proprietary methods to label disks, so the plan is to use linux's built in software raid drivers, which are more transparent and portable. As such, the disks only need to be connected to a standard SATA port, everything else is done in software. However, most motherboards don't have 12-14 SATA connectors, so a SATA controller card will be required for some of the storage array disks.

The most robust and high performance configuration would be to have two SSD drives, connected directly to the motherboard, in software raid 1, for booting and operating system, with as many 3TB drives as can fit in the case as a RAID 6 array (two drives from this array can fail without losing any data), with a hot spare (a disk already in the machine, but unused until a disk fails, used immediately to rebuild when a new disk is required to keep the array fully redundant). Assuming only 10 bays will be available to the 3TB drives, this gives capacity of 7 disks, or 21TB of raw space for storage. Without the hot spare, this increases to 24TB, and assuming we can put the SSDs somewhere other than a drive bay, it could give 27 or 30TB of raw space (depending on the hot spare).

Alternate configurations include using LVM to put the root filesystem on the same array as the main storage, with the drawback that an event that takes the storage partition down also takes the entire operating system down (such as failure of the controller card), but allows us to use 12 3TB drives with no separate boot drives, for 27 or 30TB of raw space, with no real performance hit.

A configuration that would keep all disks the same without sacrificing performance or bootability in the event of the storage array going down would be to use 2 3TB disks in a separate software raid 1 from the motherboard, for 21 or 24TB again.

One of the concerns for a raid 6 array on such a large amount of storage is that it may take a long time to build or rebuild, that is, to make the array able to suffer 2 disk failures without losing data. This must be done when the array is first created, or a disk is replaced (including being "replaced" by the hot spare). There is another more exotic setup that would alleviate this concern, though not without some drawbacks. Instead, the array could be a raid 1+0 array, where first the disks are paired, and each pair contains a perfect copy of the other in the pair, and then the data is written across these disk pairs with blocks being written to different pairs in a round robin fashion. Because all of the redundancy is simply in disks being mirror images, it is relatively fast to build or rebuild the array. This setup is currently the highest general performance raid setup, but you only get half the space, in our case, 15TB for 10 disks, 18TB for 12 (depending again on boot drive setup). The other main drawback is that failure of two drives in the same pair will cause data loss, so 2 failed disks has a chance to cause the array to become irretrievable (raid 6 requires 3 disks lost before it is irretrievable, but absolutely any 3 lost causes this, while raid 1+0 can tolerate one lost from each pair, if you are lucky).

If additional storage is required in the future, an external enclosure for drives with an eSATA interface with port multiplier may be the best option, such as this. Such an enclosure will require only 1 PCIe expansion slot, and we could use LVM2 to combine the new and old arrays into one large storage partition, despite them being disparate storage sizes. Another possible option is a USB 3.0 enclosure, but in either case, we would likely want the enclosure to report all its disks separately, and use software raid again.

Backup plan

The plan is to build two of these for a very specific reason: in case one somehow manages to die, even temporarily, the other should keep a copy of everything on it. Additionally, the plan is to keep old files on the backup server after they are deleted from the main server for a period of time, a daily job will copy new files from the entire volume, and a weekly (or longer) job will take care of removing old files from the backup server.

Backup between the two servers may be accomplished over a separate connection from the WUSTL network, if desired, and the servers are located in the same lab. The upgrade to gigabit network may make this irrelevant, depending on whether rsync can saturate the network with the read/write throughput of the arrays. If they can saturate the network, having a separate connection would leave the machines more responsive to their WUSTL network during backup, especially for operations not involving the main storage array.

Miscellaneous Notes

mdadm metadata v0.90 may be required to be able to boot from an array (the /boot partition), but limits devices to 2TB size, v1.2 should probably be used on the main array.

Ubuntu default install doesn't have mdadm installed, but the live system or the alternate install CD seems to have the needed tools.

ext4 with default blocksize has maximum size of 16TB, need to use xfs or a larger blocksize.

Ubuntu's udev may rename block devices if rebooted after a device failure, this should be tested to see if it poses a problem.

Personal tools
Sums Database