Dedicated servers

GPU servers and accelerator diagnostics

Inspect GPU hardware, compatibility, telemetry and supported sharing without changing live device configuration.

10 practical answersClear prerequisites and verification
10 guides
GPU servers and accelerator diagnostics

How do I identify the GPUs installed in a dedicated server?

Inventory dedicated-server GPUs by PCI identity and driver visibility, and distinguish an absent utility from an absent or unavailable device.

Read the answer
GPU servers and accelerator diagnostics

How do I check NVIDIA driver and CUDA compatibility on a dedicated server?

Check a dedicated GPU server’s driver, toolkit and application compatibility without confusing the CUDA capability shown by nvidia-smi.

Read the answer
GPU servers and accelerator diagnostics

How do I check GPU memory and utilisation on a dedicated server?

Inspect dedicated-server GPU memory and utilisation, understand point-in-time readings and distinguish memory reservation from active computation.

Read the answer
GPU servers and accelerator diagnostics

How do I check GPU temperature and power limits on a dedicated server?

Inspect dedicated-server GPU temperature, power, clocks and performance limits without changing cooling policy, power caps or device settings.

Read the answer
GPU servers and accelerator diagnostics

How do I identify which process is using a dedicated server GPU?

Identify GPU-using processes on a dedicated Linux server and map their process IDs to owners without killing jobs or exposing full command arguments.

Read the answer
GPU servers and accelerator diagnostics

How do I investigate GPU ECC errors or Xid events on a dedicated server?

Collect dedicated-server GPU ECC state and bounded Xid log evidence, then match the exact event to vendor guidance before resetting hardware.

Read the answer
GPU servers and accelerator diagnostics

Why can the dedicated server see a GPU but my container cannot?

Compare host and existing-container GPU visibility on a dedicated server, including driver compatibility, toolkit configuration and device assignment.

Read the answer
GPU servers and accelerator diagnostics

Can I share a dedicated server GPU using MIG or passthrough?

Compare whole-GPU assignment, MIG and supported virtual GPU options using exact hardware, hypervisor, driver and workload requirements.

Read the answer
GPU servers and accelerator diagnostics

How do I inspect GPU topology and PCIe links on a dedicated server?

Inspect dedicated-server GPU interconnect topology and reported PCIe links, and interpret idle link changes before diagnosing a hardware bottleneck.

Read the answer
GPU servers and accelerator diagnostics

How do I check AMD ROCm and CPU architecture support on a dedicated server?

Check AMD GPU, ROCm, operating-system and CPU architecture compatibility using current support matrices and installed read-only inventory tools.

Read the answer