c3

You posted contradictory information. That is arguing. Start your own post for your education. This thread was about using 4 Port 1Gb/s or 10 Gb/s NIC. I feel the OP deserves a real answer, not guesses.

Several months? Really? The OP's estimate was months. So, even with parity disabled, yes the estimate is probably still valid. My experience with Crashplan does not allow me to change that estimate. On the other hand, "out of the box", if the source(s) can write to the reiserFS directly (SATA), perhaps in parallel, the time required will be reduced.

What's to rethink? With parity calculations going on, writing directly to the actual array is never going to be that quick. That's all I was saying. The cache drive isn't really a member of the array to my way of thinking, as it isn't involved with parity, just a dumb disk to write to. But please feel free to educate me further, I welcome new learning experiences. ...and how about you be a little less abrasive and full of yourself? I never claimed to know 100% what I'm talking about, I'm simply presenting what I have set up and what my system claims to achieve in terms of speed etc. Note that I even said 'I GUESS it's capable, in theory'. That's what people do when they see something happen, and not having the full detail, make an assumption based on what they have seen. Thank you for presenting the facts above - it will certainly make me do some more research in an effort to understand this better. Please work on your delivery though. I simply stated that unRAID sustained write performance is not likely to approach a single 1Gbe interface. You decided you knew better and put up some yet to be explained numbers about a completely different situation. You don't like facts, don't blame me. 1) Explain how you can claim 180MB/sec. You have described the equipment and software. Don't the words "IP hash" give you a hint? 2) What theory are you guessing? unRAID is capable of ? Are you guessing a theory for RAID calculations based on your example of cache drive performance? The performance of COW vs XOR on striped and non-striped datasets? Is it really the calculation that determines unRAID performance or is disk access overhead a bigger factor? I think you are full of yourself. I provided information and documentation against your claims.

Sure buy a 10TB cache drive since Right Peter? OK, since you want to load 10TB, disable parity and you can get better performance. You are still looking at several months. out of the box, you can access each drive separately and fill them (reiserFS). What is the source?

Yes. I use 10 drive raidz2, SSD (10GB ZIL, rest L2ARC), compression, and replication too. Dedup seems too memory intensive for VM.

Bingo! I use unRAID. I like it. But I would not invest in 10Gbe infrastructure for an unRAID platform (yet$). Nor would I use link aggregation for a home network, the stream count/address space is too small. I have tried. However there is a 10Gbe switch (without buffering) due soon at a very reasonable price, Romley.

So, the sustained write is well under 1Gbe. Seems not so obvious, as you keep trying to hide the fact that unRAID will not sustain writes near 1Gbe. First you say it wont, then you post a write to cache image. Maybe you should rethink. IP hash would prevent a single stream from using multiple interfaces. http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1004048 I realize most of you don't do large scale storage or networking, but don't make stuff up.

Writes to cache are not sustained rates. If you which wish to see the sustained rate use a test exceeding cache size by a factor of 2. Read performance should exceed write performance.

unRAID supports a single ethernet interface. You are on your own for anything else. Remember unRAID sustained write performance is not likely to approach a single 1Gbe interface.

Just to clarify, NFS for datastore is excellent, but unRAID is not a top performing NFS platform with full random read/write IO.

It is rather hard to specify a power supply without drive count and type. You'll find the Seasonic X-750W used in many similar builds. I would recommend that if your drive count warrants it. The other factor is power supply rated efficiency, and you'll want that to be in the upper ranges. The Seasonic X series is 80 Plus Gold, only thing better is Platinum. The big power supply thread has drives at 3amps each. That's 36W each, so for all 24 drives, 864W, you'll need bigger (or so they say).

Under the CPU. http://lime-technology.com/forum/index.php?topic=17135.msg155660#msg155660 Common errors are RAM and Power Supply. Since all your components should work, something is broke. Broke RAM, error beeps. You're silent. Broke PowerSupply, no lights. You got lights. Not much else left...

I'll bet on the bent pin.

No reason to go with a Xeon for just unRAID. The X9SC? series supports i3 processors as well, but even that would not get close to the ATOM price. You'll be very happy with your X7SPA-HF-D525. Quite a few people using it and some very nice scripts. The Xeon is worth the price when you are going to build a virtualization platform like Atlas, but then we typically see the E3-1230 or E3-1240 with 8 threads.

Here's what I did with "new" motherboard. 1) connect power supply and LAN cables 2) switch on power supply 3) wait, yes, wait. The motherboard boots, BMC/IPMI gets running, gets an IP from DHCP, etc. 4) after at least 1 minute, maybe 2 minutes, hit the power button. The sound is more of a chirp than beep. You can just use a web browser to the IPMI address, IPMIview is not required.

This is normal. You can access and control via IPMIview.

YES

$/GB if the 7200rpm is cheapest get it. unRAID has spin down!

Because you should be purchasing drives for $/GB. unRAID has done much work to allow just this. The drives do not need to be the same size. This drive is $0.048/GB and the 2TB posted today as well is $0.050/GB. However each case can include things like refurbished, 90 day warranty (like this one for $120). Personally I am looking for $0.04/GB...

Ditto. Is the standard cooler the way to go? If not, which CPU coolers are compatible with the X9CM? I use the Intel provided heatsink and fan. John's pictures indicate the same.

I wish everyone would read this. I have been warning that checking a serial to validate warranty just says the entire unit is under warranty. An external enclosure has the same serial number as the drive inside, but the manufacturer does know it was not a bare drive. As warranty period drops, this is less and less an issue, but it is nice to know you are giving up the warranty when you break that seal. Don't live in denial that you are going to get away with fraud.

not if you use udev rules.

vmkfstools –i disk0.vmdk –d thin disk0-thin.vmdk

As John said, it depends on what will be running on those guests. unRAID with no addons is easy in 1G, but could be much more. You'll have to decide how to divide up the RAM, 1+3+3, 2+2+3, remembering to save about 800M for ESXi. Also keep the number of vCPU as low as possible, 1 is best. Be sure to install vmware tools on the guest, it will help with memory management.