Only 8x8GB since I'm working on CPU limited simulation tasks. One of the CPUs and 4 of the DIMMs is on a daughter board in the Z640, which makes the whole thing even more impressive. It's a fairly compact machine. Storage is just a single SSD and single HDD. Graphics is a low end Quadro just to get the 4x mini-DisplayPort. The HP Z8xx series would probably be a better choice if a person is going to go crazy on the rest of the hardware. But that's just conjecture on my part based on the fact that my current processors are running near TDP (but aren't being throttled down based on the monitoring I have done.)
I have a z820 32 cores, 128GB of RAM, 3 nVidia GeForce GT 710 cards, and 3 x 5TB drives. You're not missing much with the z640. I have a z620 as well. The z620 just has 2 fewer PCI-E slots. The difference is the same for the z840 and z640.
Can you explain the 3x Nvidia 710s? The only reasoning I can think of is running ~6 monitors silently, but the heatsink-only cards aren't usually used in prebuilts so I'm a little confused.
I just assume that the larger Z8 chassis might handle the heat a little better? The Z6 works for my use, but the CPUs are definitely running on the hot side.
Maybe. With a bunk front fan on my z820, everything still runs relatively cool at maybe a max of ~55C (2 CPUs, 3 GPUs, 16 sticks of RAM) according to sensors util in RHEL.
Electric grid dynamic/transient event (disturbance) simulations. It's essentially modeling the response of electromechanical equipment (generators and large loads), large solid state devices (utility scale renewables and fast-switching reactive devices), lumped load models, and protective devices to events on the electric system. The simulations run in ~1ms timesteps over 10-30 seconds of real world time (which can take 4-20 minutes of CPU time). Each event is independent and they parallelize pretty much linearly. Python multiprocessing is the lifeblood of this work.
Edit: I guess I should be clear and say that the simulations themselves are running on a commercial software package (Fortran based) and I'm simply setting them up, spinning them off, and then processing all of the resulting data in Python.
Neat. I do a lot of RF simulations. Even though it's 60 Hz vs 60 GHz, I suppose a massive electrical grid has the same distributed effects that can cause standing waves and stability issues.
