8 ms·
I've got an HP Z640 workstation with dual E5-2690 v4's and it is basically silent under full load. It has no problem running all 28 cores at 3.2GHz which is th
by the_real_sparky 9y ago
I've got an HP Z640 workstation with dual E5-2690 v4's and it is basically silent under full load. It has no problem running all 28 cores at 3.2GHz which is the max all-core speed for that processor.
- phren0logy 9y agoThat's really impressive. How many DIMMS and spinning HDDs is it cooling?
- the_real_sparky 9y agoOnly 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.)
- madengr 9y agoWhat'ya simulating?
- oso2k 9y agoI use my z820 and z620 for running OpenShift/Kubernetes/Containers/VM workloads.
- the_real_sparky 9y agoElectric 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.
- madengr 9y agoNeat. 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.
- oso2k 9y agoI 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.
- the_real_sparky 9y agoI 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.
- oso2k 9y agoMaybe. 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.
- PascLeRasc 9y agoCan 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.
- oso2k 9y agoExactly. 6 monitors in 3x2, they're super cost effective at $40-$45, and silent as you mentioned.
- closeparen 9y agoWho needs something like this? What is your job? What do you use it for? Really curious. I deploy code to a large production server cluster, my friends in academia submit code to large scientific computing clusters, but I don't know of anyone with this much power in their own desktop. I guess I've been in video edit suites with machines much more powerful than the average PC, but not this crazy.
- gh02t 9y agoYou see workstations like this in what I'd call "interactive simulation." Stuff where you're designing something complex via simulation and want to be able to rapidly evaluate changes. For example ANSYS (hybrid finite element simulation and CAD) works best on a single workstation and you want it to be fast so you can make quick changes. Or when your geometry is so complex it doesn't fit in any reasonable amount of memory and your code doesn't support distributing it (which is very hard for a lot of simulation types). The (Sauber? I think HP sponsors them) F1 car from the image in the article is a good example of someone who might one one of these. A lot of battle-tested engineering software doesn't do distributed-memory parallelism well. I've personally used MCNP and GEANT4 (particle physics software) on a 48 core, 1 TB RAM workstation. It's interesting browsing the web while the computer is running gigantic calculations in the background on two [physical] CPUs, hooray for the process scheduler doing a good job.
- qq66 9y agoUsually, engineering simulations (finite element analysis, etc)
- the_evacuator 9y agoMay I suggest compiling software? For people without access to distributed build and test a box like this is really nice. I do have access to a huge distributed build farm but even then I use a beefy HP workstation because the software that drives the build farm is itself a bit of a pig, and massively multithreaded.
- oso2k 9y agoI'm an OpenShift Consultant for Red Hat. Super awesome to be able to run or build from scratch a 9 node HA OpenShift cluster without incurring an AWS or other infrastructure costs.
- blinkingled 9y agoSeconded - HP Z series are very good quality workstations - very less noise and great Linux compatibility. You can find them cheap as HP refurbs if you want warranty like new, otherwise eBay has them for quite cheap too.