Without me having the time and the hardware to do a real world test this week I’ve been spending a lot of time reading up on my future setup.  It would appear that since GlusterFS doesn’t (at this time but Raid6 is slotted for version 1.7) support parity blocks, a Raid5 style setup would not be feasable.  The best you could hope for with as little hardware as I have is to run a Raid 0 or Raid 1 setup.  With some advanced setup and multiple drives across 2 computers, a Raid01 setup would be possible, but I don’t have the means to try it out.

Once I get all the kinks out of my current, sloppy setup, I will post a howto describing both the installation, setup and pitfalls I’ve encountered along the way.

Now the setup I have experimented with was just a means for me to see what would happen if I setup 2 machines, one with a 200GB partition and the other with a 60GB partition.  After I configured both machines and mounted the GlusterFS clustered/striped file system to /mnt/pool, I had a nice, shiny ~250GB partition.  This is cool for 3 reasons.

1. I don’t need the same hardware in both machines to make this setup work.

2. I get approx 100% of both drives being used in the setup.  HD1 + HD2.

3. I get a slight speed increase due to the striping, similar to RAID 0.  When data is written or read from the drives, I have 2 sources instead of one.  So my maximum theoretical bandwidth is no longer limited to hard drive transfer rates, but to the 2 system’s drives or the network, whichever comes first.

I am testing these out on a slower network setup, 100Mb/s bandwidth, but once I have a more advanced setup I want to move to something a little more robust, 1000Mb/s.

My setup right now has one MAJOR flaw in it.  If one drive goes down (drive, system, network) then the whole /mnt/pool file system is unavailable.  Again, think Raid 0.  For me this would only be useful if I had a bunch of drives that I just wanted to use as a faster way to serve unimportant data.

Since all the data I have is important, that leaves me with 2 options.  Simple mirroring across multiple drives (Raid 1) or what I’m going to try in my next step: File replication over striping.  Think more like Raid 5.  Benefits will be the following:

1. Redundant data.  Data is mirrored across multiple drives to allow redundancy and a speed increase.

2. Speed increase.  Since data is stored on more than one drive, the machines read from multiple locations and faster read times can be achieved.

3. Storage space.  While not as impressive as Raid 0 on it’s use of space, Raid 5 uses a minimum of 3 drives and stores like this.

Drive one:     1/2/3

Drive two:     2/3/1

Drive three:   3/1/2

Data is figured being “Size * (Number of disks – 1) / number of disks).  So assuming these are three 1TB drives, it would look like this:  3,000,000,000 * (3 – 1 = 2) / 3 = 2,000,000,000.  So you get storage space equal to ~66% of the total drive space in the above example.  Three 1TB drives will give you 2TB of usable space.  Three 100GB drives give you 200GB usable.  Assuming you use 5 drives…5,000,000,000 * (5-1=4)/5 = 4TB.  3 gives you 2, 5 gives you 4, 10 gives you 9.

4. If one node (computer/network connection/hard drive) goes down, the whole system keeps on going.  Again, since the same data is stored on multiple drives, if one system goes down, the controlling machine or machines will see this and know to pull data from the backup source.

Looking at what I had listed earlier, if we take down drive 1, the data on drive 1 is also on drive 3.  So drive 3 takes over serving drive 3 and 1 data and drive 2 keeps on going since it’s the original.

My main problem I have right now is finding another system that has a decent sized drive in it.  I found a spare machine here with a nice 13GB drive in it, I know, it’s huge.  But for real world tests, this is too small.  I want at least an 80GB drive and while I had 2 not so long ago, I gave them both to my father in law because I saw no need to keep such small drives.  I could use them now.

More to come when I have it.  Stay tuned…

I wound up finding GlusterFS that suited my needs. It allows you to span hard drives across multiple networked systems and mount all machines as 1 drive. Now I’m suddenly finding a possible use for all of those old systems around the house.

The only limit to how big you go is how many machines you have linked together and how fast your network is. You’ll need a fast backbone to pull GB/sec but I have seen places on their site make reference to some really impressive numbers.

I’m at the researching stage at this point and garhering materials, so as soon as I have a working model I will post a howto on my setup and any issues I run into.

The official release should be announced sometime in October (hence the “10″ in the 8.10).  After all of the other releases I’ve upgraded to in the past, there have sometimes been subtle, yet important changes to the OS and here recently, major changes to the system to make it both more user friendly and a more robust system. I never get bored with the new releases.

Their countdown script doesn’t seem to work for non-LTS versions (April releases) so the image isn’t updating as of this posting. Sucks but what do you do?

Edit: It started working when the official release date was announced.