Help with building a number cruncher

Associate
Joined
2 May 2013
Posts
17
Hi all,

You seem like a knowledgeable forum - and I was hoping you could help me out.

I'm working on some research, that research requires doing a lot of number crunching on ~55GB of data. To make this process as quick as possible, all that data should be in memory. A vast majority of the work it's doing is on floating point numbers. This'll be an iterative process running over several weeks.

I've built a few machines before, although only AMD. I thinking an Intel CPU might be better for this job.

Graphics are not important, and it'll only be dumping a few GB of results to disk every 24 hours, so I'm not that fussed about the drive either. As an afterthought - this machine might eventually become a development workstation once it's done number crunching.

I'd like to spend as little as possible (doesn't everyone), ideally <£1000. Putting together a few pieces as a starting point (And I'm already over budget):

Intel Core i7-3820 3.60GHz (Sandybridge-E) Socket LGA2011 Processor - Retail
2x Corsair Vengeance RED 32GB (4x8GB) DDR3 PC3-14900C10 1866MHz Dual/Quad Channel Kit (CMZ32GX3M4X1866C10R)
Asus P9X79 Intel X79 (Socket 2011) DDR3 Motherboard
Western Digital Caviar Green 1TB SATA 6Gb/s 64MB Cache - OEM (WD10EZRX) **SINGLE PLATTER** HDD
Cooler Master Elite Power 500W Power Supply
Xigmatek Asgard Midi Tower Case - Black
Sapphire HD 6450 HM 512MB GDDR3 PCI-Express Graphics Card (11190-04-20G)
Microsoft Windows 7 Professional SP1 64-Bit - OEM (FQC-04649)
£1,159.91

Thanks in advance!
 
What sort of number crunching? Your code or someone else's? Windows seems a strange choice if it's your own code, so Ansys et al seems likely.

A 55gb fluid dynamics case would take a long time to grind through, solid mech not so bad. Serious number crunching is memory bandwidth limited. Socket 2011 with fast, low latency ram is significantly better than Ivy - just comes down to dual/quad channel ram. 4/6 cores is tricky, I'm not sure whether four cores is sufficient to saturate the quad channel memory bandwidth or not. Probably code dependent.

Your basic premise - socket 2011, 64gb ram looks about right. SSD is worth nothing if the dataset fits in ram. The bang/buck approach is multiple cheap computers networked together however. Currently a pile of ivy bridge systems, no gpu, no hard drive, network boot. That's the road I'm planning on going down in a few months time - small cluster instead of one big computer.
 
Thanks for the feedback.

@CinderzFTW - I'm pretty certain that replacing the 64GB RAM with an SSD would slow it down massively. I know SSDs are quick, but not that quick. Especially when being read and written by multiple threads.

@JonJ678 - It's crunching the data to produce and adapt a decision tree. All inputs are floats. The reason I'm going with windows is because a majority of the libraries that handle the data are already written in C# + that's the language I know best.

Multiple machines sounds like an interesting idea, especially as I could start with one and expand as needed.

Anyone got any ideas on recommended hardware? Any adaptations to what I've got listed above?

Cheers.
 
I have non-hyperthreaded, overclocked quad cores planned. Right now thats the 3570k iirc, about £175 at last check. Gigabyte m-atx board, samsung green ram. Haswell is out in a month if you're patient.

C# support pthreads or openmp? May have trouble with standard libraries' thread safety. The multiple machine version is mpi, e.g. openmpi. Usual keyword is beowulf, but it's normally built using linux.

Whether clusters work well depends on the problem. If it parallelises easily they're great, if someone is willing to build and maintain the thing.
 
I'm re-evaluating my build. Any thoughts on the i7 3930K over the 3820? Less speed, 50% more threads.

If I have 12 threads making memory requests, is the RAM likely to be a bottleneck? Or am I likely to be able to squeeze, say a 25% performance boost?

I know this is all code dependent as well as hardware dependant. Advice would be appreciated all the same.
 
I'm re-evaluating my build. Any thoughts on the i7 3930K over the 3820? Less speed, 50% more threads.

If I have 12 threads making memory requests, is the RAM likely to be a bottleneck? Or am I likely to be able to squeeze, say a 25% performance boost?

I know this is all code dependent as well as hardware dependant. Advice would be appreciated all the same.

If you are doing really memory intensive tasks, i.e. intensive calculations, the bottleneck will likely be on the clock speed of the RAM if you are choosing between 3820 or 3930K.
The only downside of 3820 is that you need a fairly nice motherboard to be able to overclock it, but you will get one anyways.

So, I would really consider getting R4E and 3820 or the next gen R5E and 4820(?) if you can wait for IB-E.
Probably 2400MHz RAMs will do the job for you... but the only three of these will cost you ~£1000, then you still have to buy a graphics card as SB-E do not have an onboard graphics.
 
YOUR BASKET
1 x Intel Core i7-3820 3.60GHz (Sandybridge-E) Socket LGA2011 Processor - OEM £239.99
1 x Asus P9X79 Intel X79 (Socket 2011) DDR3 Motherboard £199.99
2 x Avexir MPower Yellow Series 32GB (4x8GB) DDR3 PC3-12800C10 1600MHz Quad Channel Memory Kit (AVD3U16001008G-4CM) £179.99 (£359.98)
1 x SanDisk Extreme SSD 120GB 2.5" SATA 6Gb/s Solid State Hard Drive - (SDSSDX-120G-G25) £83.99
1 x Corsair Carbide 300R Mid Tower Case - Black £64.99
1 x Alpenföhn K2 Mount Doom CPU Cooler (Socket 775 / 1155 / 1156 / 1366 / 2011 / AM2 / AM2+ / AM3 / FM1 / FM2) £56.39
1 x OCZ ZS Series 550W '80 Plus Bronze' Power Supply £54.98
Total : £1,075.31 (includes shipping : £12.50).



i7-8320 with 64GB of RAM
Big cooler which should help you get a decent overclock and get the job done much faster
and an SSD

What more could you ask for?

EDIT: your gonna want a cheap graphics card too so add £20-£30 on to that
 
Last edited:
Thanks for that.

@deFiniLoGy - Too many acronyms for a novice. R4E? IB-E? SB-E? Sorry, stuff I should probably know. So you reckon it'd be worth me splashing out on Kingston HyperX Beast 32GB (4x8GB) PC3-19200C11 2400MHz?

@MonsterMoshi I have 2 concerns with your build. You've downgraded to a slower RAM. Any reason other than price saving? Also, I'd need more hard drive space for dumping results, and don't really need the speed of an SSD, as once everything is in RAM it shouldn't have to touch the drive for 24 hours. Nice idea with the OEM CPU and chunky cooler. :-)

Anyone got any experience of the P9X79 in it's various forms? Any other motherboard I should be considering?
 
I have no idea what he means bu R4E?
But IB-E and SB-E are the 2011 socket versions of SandyBridge and IvyBridge
IvyBridge-E is yet to be released

Faster ram makes things a hell of a lot more expensive
I don't know anything about number crunching but faster ram makes little to no difference in general use and gaming
The RAM i selected has decent heatsinks and perhaps with an extra cooler there is no reason you couldn't overclock it it 1866 or maybe even 2133
YOUR BASKET
1 x Arctic RC Turbo Module PWM Memory Fan £8.99




I always like to recommend an SSD at least for use as a workstation as per your plans as it really does make a huge difference but that is down to you
 
Last edited:
Going off on a tangent here. If I was to go for the 3820, is that definitely worth the extra money over the AMD FX-8350? Benchmarks generally seem to put the two pretty much on a par. I can't find and gflops numbers anywhere though - which is what I'd really be interested in. Anyone know where I could find accurate comparisons of those? What could I get them to if OCed?

Cheers.
 
Use a 3930k, far better for number crunching than a 3820 and also easier to overclock.

R4E is short for Asus Rampage IV Extreme, probably the best mobo you can get for X79 but for just number crunching you should be able to use something a little cheaper.
 
Last edited:
Going off on a tangent here. If I was to go for the 3820, is that definitely worth the extra money over the AMD FX-8350? Benchmarks generally seem to put the two pretty much on a par. I can't find and gflops numbers anywhere though - which is what I'd really be interested in. Anyone know where I could find accurate comparisons of those? What could I get them to if OCed?

Cheers.

http://cpuboss.com/cpus/Intel-Core-i7-3820-vs-AMD-FX-8350

http://techreport.com/review/23750/amd-fx-8350-processor-reviewed/3

Found these comparisons, but nothing to compare Gflops.
 
Thanks, the second one is very interesting, especially the FPU tests. Summed up nicely in this paragraph.

The FX-8350 takes the top spot in the CPU Hash test, not surprising given the relatively strong performance of the AMD processors in this integer-focused benchmark. The more FPU-intensive fractal tests are a very different story, with the Sandy and Ivy Bridge-based chips topping the charts.
Sounds like the Intel chip wins.
 
Thanks, the second one is very interesting, especially the FPU tests. Summed up nicely in this paragraph.


Sounds like the Intel chip wins.

The 3930k I mentioned earlier performs almost the same as the 3960x in those benches. If you run the two @the same clock speed it is hard to tell them apart.
 
The 3930k I mentioned earlier performs almost the same as the 3960x in those benches. If you run the two @the same clock speed it is hard to tell them apart.
The 3930k is a serious consideration, but having seen the tech report link that Idleman posted - the 3930k and the 3820 are pretty much tied on the FPU tests, both are annihilated by the 3960x. I'm not sure exactly what those tests incorporate, but that sounds like the most suitable test for me to compare with.

So having seen that I'm tempted to stick with the 3820.

Edit: Sorry, I wrote a load of rubbish. Just noticed the 3930k isn't mentioned in that test, it's the 3770K.

Edit: Quick bit of research. For those that are interested - Looks like the Julia FPU test runs with a 32bit FP datatype. (Most suitable for me). The Mandel test runs with 64bit FP datatypes.
 
Last edited:
I'm looking at using the 3930K overclocked with the Corsair Hydro H55

I've never used any form of liquid cooler before. Is this a good idea? Any other tips? I guess I should have an extra case fan to help move air through the case?

Thanks.
 
Back
Top Bottom