Discussing multi-user systems involves understanding how multiple users access and share resources simultaneously.
Discussing multi-user systems involves understanding how multiple users access and share resources simultaneously.
I build a highly capable overkill PC. I link several stations, each with a monitor, mice, keyboard, and USB hub. Multiple users can connect and collaborate simultaneously from different locations. The challenge is: how do I achieve this? The solution must run on Windows since these are advanced 3D design tools that only support the platform.
Check out the Wikipedia page on multiseat configurations and the Reddit discussion about it. It's not unusual in my country, but the setup tends to be more complex. I've noticed such arrangements in places like McDonald's. The main issues are: resolution caps at 800x600 or 1024x800, only changes from the main server, no support for 3D hardware acceleration—so no gaming or rendering, just useful for running client-side workstations except with multisheet/document tasks. There have been improvements with newer Intel and AMD solutions that use low power (5-20W) for mid-range use, but these setups are rarely viable anymore. Many people now opt to use tablets or phones at public locations instead. You can try a free trial here: http://www.miniframe.com/products/softxp...Technology
My responses are quite general. If you mean updating the same document on various connected devices, it feels like cloud computing. It helps streamline tasks like documents or presentations, but working together in 3D environments can be slow and hard to manage. Each person usually focuses on one task to avoid confusion later. When something fails, you review what went wrong and fix it. If you're building a system for rendering, setting up a dedicated farm with powerful machines makes more sense than using a single overpowered PC. Using multiple less powerful machines together is still an option but isn't as common these days.
For a professional setup, a VM cluster is likely the best choice. A single system equipped with two E5 processors and 128-256GB of RAM running HyperV can serve as a base. Smaller machines connect via RDP to access virtual Windows 7/8 Pro environments, each allocated specific resources like 4 cores and 16-32GB of RAM. This approach can also be implemented using thin clients based on the requirements.
IOMMU, definitely feasible so. Start with step 1) verify your hardware supports AMD-Vi or Intel VT-d – if it does, it’s likely already enabled. For Intel users, proceed to step 2) switch away from Windows as the primary OS; it’s time to explore Linux. In step 3) install libvirt, QEMU-KVM, and Virt-Manager. At step 4) configure your Linux host with PCI-Stub support. Make sure multiple GPUs are recognized by the PCI-Stub. Step 5) build a Windows VM, configure it, and run it. Then clone the setup. In step 6) add peripherals like GPU, keyboard, mouse. Disable VNC and VGA displays. By step 7) launch the VM, it should function independently with its own keyboard, mouse, GPU, etc. Basic guidance, but you’ve got it. Multiple GPUs running VMs on one power unit is possible.