Sign in with
Sign up | Sign in

Hardware Component Selection For Each Node

How To: Building Your Own Render Farm
By

Processor selection is dependent on two things: how much you're able to spend and resulting value. Even more so than normal desktop systems, render nodes are sensitive to these factors. There is a certain point on the processor performance curve where the next increment in compute speed results in a pricing jump that is out of order with the increase in performance. However, with today's higher bus speeds, too little cache on a processor can leave it starved for data. So, for discussion's sake, we'll shy away from Celerons as the bottom-end processor and instead use low-end dual-core Pentiums.

Using a lower-cost dual-core Pentium can give you a very inexpensive render node with a minimum sacrifice of performance. The Cinema 4D Release 10 chart shows the performance curve. While dual-core Pentiums and Celerons are not in these charts, you can see that there are spots in the Intel processor lineup where there is a significant jump in price for very little increase in performance. Find the price that you are willing to pay and then look at the best performance for your price range.

A dual-core Pentium, a slightly faster Core 2 Duo, or even a low-end Core 2 Quad are all good choices for building a node. It's really all about how much you want to spend here, because this is the single most expensive component required for each node. At the other end of the spectrum, building a Core i7-based system is going to increase the price of the entire node out of proportion to its increased performance. However, if you want to moderately future-proof your nodes and increase the likelihood of being able to get a processor upgrade for it later, you may want to consider that route.

A low-profile 1U heat sink is required for a corresponding case, most of which are made to specifically draw air from the front to the back of the machine. The limits in choices in processor cooling also limit (or even restrict) the possibility of overclocking the nodes. The Dynatron P199 pictured above is fairly typical and can handle quad-core processors.

Storage

Because the local drive will just be used for your operating system and applications, pick a small drive. An 80 GB drive is the smallest SATA drive commonly available, so that's what we recommend. However, if you use a drive with a digital audio workstation (that setup is discussed below) then a 320 GB, or maybe even a 500 GB drive, is a better choice.

Assuming you are building multiple nodes, you are only going to buy a single DVD drive (even if you are building 10 systems). You'll build your first machine, install the operating system plus software with updates, and then use Sysprep to blank the registration number. Next, use Norton Ghost 14 or similar drive-imaging software to clone the disk for subsequent machines. Each new system will require a fresh network name, and you will have to enter the Windows license key for each of the cloned machines. This keeps you from having to install the operating system, the applications, and the updates on each of the machines separately. If you build additional nodes using the same hardware in the future, it's also a good idea to clone one extra drive and keep the original as a master drive.

The operating system for your nodes should be similar (but not necessarily identical) to your operating system. Since 64-bit Windows XP is still available through OEM channels, there is no need for you to even consider putting Vista on your nodes, wasting memory and processing power. If your 3D application's network renderer (and all of the plugins that your network renderer needs, including third-party plugins) are supported under Linux, you may want to consider running Linux on the nodes instead, which makes the nodes even cheaper. A total of 10 copies of XP (for 10 nodes) may sound like a big expense, but it actually adds $140 per unit, pushing the cost of these machines to about $485 per unit for a dual-core node or $610 per unit for a quad-core configuration.

Component
Dual-Core
Price
Quad-Core
Price
Casing
Supermicro CSE-512L-260
$94.99
Supermicro CSE-512L-260
$94.99
Motherboard
Asus P5B-VM SE$59.99
Asus P5B-VM SE$59.99
Processor
Intel Core 2 Duo E7200$91.99
Intel Core 2 Quad Q8200
$164.99
Heat Sink
Dynatron P199$30.99
Dynatron P199$30.99
Memory
Patriot Viper PVS24G6400LLK – (2 GB x 2)$51.99
Two Patriot Viper PVS24G6400LLK – (2 GB x 4)$103.98
Hard Drive
Seagate Barracuda 7200.10 ST380215A 80 GB$36.99
Seagate Barracuda 7200.10 ST380215A 80 GB$36.99
Operating System
Windows XP Pro 64-bit$139.99
Windows XP Pro 64-bit$139.99
Total (With Shipping)

$485.11

$610.10


So here’s what we’re looking at for a per-node price, with both dual- and quad-core processors. Remember, these are prices we found when this article was written, and they should change in the near future.

Display all 117 comments.
Top Comments
  • 13 Hide
    MonsterCookie , July 17, 2009 1:33 PM
    Due to my job I work on parallel computers every day.
    I got to say: building a cheapo C2D might be OK, but still it is better nowadays to buy cheap C2Q instead, because the price/performance ratio of the machine is considerably better.
    However, please DO NOT spend more than 30% of you money on useless M$ products.
    Be serious, and keep cheap things cheap, and spend your hard earned money on a better machine or on your wife/kids/bear instead.
    Use linux, solaris, whatsoever ...
    Better performance, better memory management, higher stability.
    IN FACT, most real design/3D applications run under unixoid operating systems.
Other Comments
  • 4 Hide
    Draven35 , July 17, 2009 9:14 AM
    People have been saying that for several years now, and Nvidia has killed Gelato. Every time that there has been an effort to move to GPU-based rendering, there has been a change to how things are rendered that has made it ineffective to do so.
  • 3 Hide
    borandi , July 17, 2009 9:43 AM
    With the advent of OpenCL at the tail end of the year, and given that a server farm is a centre for multiparallel processes, GPGPU rendering should be around the corner. You can't ignore the power of 1.2TFlops per PCI-E slot (if you can render efficiently enough), or 2.4TFlops per kilowatt, as opposed to 10 old Pentium Dual Cores in a rack.
  • 4 Hide
    Draven35 , July 17, 2009 9:53 AM
    Yes, but it still won't render in real time. You'll still need render time, and that means separate systems. i did not ignore that in the article, and in fact discussed GPU-based rendering and ways to prepare your nodes for that. Just because you may start rendering on a GPU, does not mean it will be in real time. TV rendering is now in high definitiion, (finished in 1080p, usually) and rendering for film is done in at least that resolution, or 2k-4k. If you think you're going to use GPU-based rendering, get boards with an x16 slot, and rsier cards, then put GPUs in the units when you start using it. Considering software development cycles, It will likely be a year before a GPGPU-based renderer made in OpenCL is available from any 3D software vendors for at least a year (i.e. SIGGRAPH 2010). Most 3D animators do not and will not develop their own renderers.
  • 0 Hide
    ytoledano , July 17, 2009 11:31 AM
    While I never rendered any 3d scenes, I did learn a lot on building a home server rack. I'm working on a project which involves combinatorial optimization and genetic algorithms - both need a lot of processing power and can be easily split to many processing units. I was surprised to see how cheap one quad core node can be.
  • 0 Hide
    Draven35 , July 17, 2009 11:39 AM
    Great, thanks- its very cool to hear someone cite another use of this type of setup. Hope you found some useful data.
  • 13 Hide
    MonsterCookie , July 17, 2009 1:33 PM
    Due to my job I work on parallel computers every day.
    I got to say: building a cheapo C2D might be OK, but still it is better nowadays to buy cheap C2Q instead, because the price/performance ratio of the machine is considerably better.
    However, please DO NOT spend more than 30% of you money on useless M$ products.
    Be serious, and keep cheap things cheap, and spend your hard earned money on a better machine or on your wife/kids/bear instead.
    Use linux, solaris, whatsoever ...
    Better performance, better memory management, higher stability.
    IN FACT, most real design/3D applications run under unixoid operating systems.
  • 0 Hide
    ricstorms , July 17, 2009 1:38 PM
    Actually I think if you look at a value analysis, AMD could actually give a decent value for the money. Get an old Phenom 9600 for $89 and build some ridiculously cheap workstations and nodes. The only thing that would kill you is power consumption, I don't think the 1st gen Phenoms were good at undervolting (of course they weren't good on a whole lot of things). Of course the Q8200 would trounce it, but Intel won't put their Quads south of $150 (not that they really need to).
  • 0 Hide
    eaclou , July 17, 2009 1:50 PM
    Thanks for doing an article on workstations -- sometimes it feels like all of the articles are only concerned with gaming.

    I'm not to the point yet where I really need a render farm, but this information might come in handy in a year or two. (and I severely doubt GPU rendering will make CPU rendering a thing of the past in 2 years)

    I look forward to future articles on workstations
    -Is there any chance of a comparison between workstation graphics cards and gaming graphics cards?
  • 1 Hide
    cah027 , July 17, 2009 1:52 PM
    I wish these software companies would get on the ball. There are consumer level software packages that will use multiple cpu cores as well as GPU all at the same time. Then someone could build a 4 socket, 6 GPU box all in one that would do the work equal to several cheap nodes!
  • -1 Hide
    sanchz , July 17, 2009 3:22 PM
    Correct me if I'm wrong, but wouldn't 30 million hours be 30,000,000/24 = 1,250,000 days which would in turn be 1,250,000 / 365 = 3,425 YEARS!!! O.o
    Please someone clarify this. How could they render a movie for 3,000 years? Did they have this render farms hidden in Egypt??
  • -2 Hide
    nemi_PC , July 17, 2009 3:27 PM
    Some thoughts for small nodes:
    1) Cases cablable of taking a 2 slot grpahics card woudl future proff setting up anode at this time in case GPU rendering does become applciable over the lifetiem of the node. So (m)ATX cases not rack mounts
    2) Resale of a (m)ATX "reglaur" looking desktop a few years down he road to "home users" is easier than a rack mount server. So should factor that into the value.
    3) With 500-1TB being the sweet spot for Gb/$ I would go with those drives and use the render node also as a distributed (redundant) back up solution , this address where are you going to store all your work over the years.
  • 2 Hide
    eyemaster , July 17, 2009 3:33 PM
    I'm with you sanchz. But I think they mean per single processor. Say, if you had a common desktop computer and tried to render the whole transformers 2 movie, it would take thousands of years. If you have 10000 processors doing the job, you can do it within a year or less.
  • 0 Hide
    mlcloud , July 17, 2009 3:33 PM
    sanchzCorrect me if I'm wrong, but wouldn't 30 million hours be 30,000,000/24 = 1,250,000 days which would in turn be 1,250,000 / 365 = 3,425 YEARS!!! O.oPlease someone clarify this. How could they render a movie for 3,000 years? Did they have this render farms hidden in Egypt??


    What do you think the meaning of parallel processing is? Doing a lot of that work at once, right? If we have a huge render farm of 5000+ processors, we cut down that time to less than a year, wouldn't we?

    Of course, a lot of that depends how fast each processor in the render farm is, but the general public won't care about that; just give 'em the huge numbers and don't tell them you were using 1.6ghz celery's in your render farms.
  • -1 Hide
    one-shot , July 17, 2009 4:10 PM
    Hmmm. The standard electrical voltage for residential dwellings (United States) on a 120/240V two phase installation is plus or minus 5% of 120V, not the 110V which is mostly stated. So 15A * 120V = 1800VA or Watts, not 15 * 110V.
  • -5 Hide
    ossie , July 17, 2009 4:41 PM
    In view of eventual future GPU offloading, at least a 3U case for 4x2slot-PCIe GPUs would be necessary, so the upgradeability of 1U cases is limited to one 1slot GPU. But such a monster would get easy over 1kW, posing more challenges for power supply and cooling (generated noise left apart).
    As MonsterCookie pointed already out, use some good scaling multi-processor/-node OS for good distributed performance (m$ doesn't apply).

    Finally a decent article on TH... almost without the usual vi$hta or $even (aka vi$hta sp2+) m$ pu$hers behind.
    What? xpire x64 is working for TH? almost unbelievable...
    Also, none of the usual m$ fankiddie and gamer comments, (at least) till now... :) 
  • 0 Hide
    dami , July 17, 2009 5:02 PM
    Another example, getting out of the computer jargon...

    If a task took 100 man hours, that means it took 2 guys 50 hours each to do something. If you did that with 10 guys, it would take each man 10 hours of work. There is a point of diminishing efficiency, which is mentioned in the article. The extreme to this is, it would take 100 men, 1 hour of work to complete the same task. The efficiency has been drasticly reduced.

    This is whats being done in these rendering farms. A bunch of processors are put together, tasked with a job, and they belt out the results. If you did that with just one processor, it would take the 3k years in egypt to come up with a result.
  • -1 Hide
    Anonymous , July 17, 2009 5:11 PM
    borandiAnd soon they'll all move to graphics cards rendering. Simple. This article for now: worthless.

    I do agree with graphics card rendering,but don't think this article is worthless!

    When I read about xeons, I also read about AMD making similar, low power processors like that (45nm or lower?,and a TDP of around 65W, which is 30W lower than their previous processor line).
    It might not be beneficial to buy xeons, but perhaps it might when going with AMD.
  • -8 Hide
    aspireonelover , July 17, 2009 5:17 PM
    I rather spend the money on helping developing countries. (I know this has nothing to do with farm render and stuff but the amount of money they've spent on these machines is an incredible amount)
    Like buy a few XO netbooks for the developing countries, and sponsor lots of children.
  • 0 Hide
    Anonymous , July 17, 2009 6:15 PM
    How about network booting the cluster - we have found it easier to manage upgrades/patches as you just need to reboot the nodes that need upgrading. Also makes each node a little cheaper and saves a fair bit on power consumption.
    Another idea we have been playing with is using cheap USB keyfobs either as system drives or to persist config data etc. - much faster boot times, very low power consumption and great MTTF.
Display more comments
React To This Article