Display Support: WDDM is required if you want to plug a monitor into your GPU.
If you’re convinced TCC is better, here is how to enable it.
TCC vs. WDDM: Which NVIDIA Driver Model Is Better For Your Workload?
| Feature | WDDM | TCC | Benefit for Compute | |---------|------|-----|---------------------| | | Enabled (2s timeout) | Disabled | Run kernels of any duration | | GPU as display device | Yes (monitor output) | No | Frees resources for compute | | Memory paging | Managed by Windows | Direct GPU memory access | Lower latency, higher bandwidth | | Process isolation | Full preemption | Minimal context switching | Higher sustained throughput | | Kernel launch overhead | High (via OS) | Low (direct to GPU) | Better for many small kernels | | Remote DMA (RDMA) | Not supported | Supported (over InfiniBand/ROCE) | Essential for multi-GPU clusters | tcc wddm better
When configuring enterprise NVIDIA GPUs—such as the A100, H100, RTX 6000 Ada, or Quadro series—you will face a critical configuration choice in the NVIDIA Control Panel or command line: selecting between the and WDDM (Windows Display Driver Model) driver models.
Despite the performance advantages of TCC, WDDM remains mandatory or superior in several key situations:
For a dedicated compute node, these downsides are irrelevant. For a hybrid workstation, use a hybrid driver setup. Display Support: WDDM is required if you want
If you attempt to connect a monitor to a GPU running in TCC mode, Windows may detect it as a "Standard VGA" device, leading to unpredictable behavior and potential system crashes requiring a full reboot.
: Users have reported that switching to TCC can increase pageable memory copy speeds by up to 50%. This makes TCC the superior choice for "big data" transfers where WDDM’s management overhead would otherwise cause a massive "speed loss". 3. Stability and "Headless" Reliability
You can run a single kernel for weeks without interruption. Furthermore, TCC allows for "Peer-to-Peer" (P2P) transfers between GPUs (NVLink) without copying memory through system RAM. WDDM often blocks direct P2P for stability reasons. WDDM: Which NVIDIA Driver Model Is Better For Your Workload
, but not a silver bullet.
要理解为什么TCC模式更优，首先需要了解这两种模式的设计初衷和工作原理。