Obtaining a Tesla Driver and CUDA Toolkit
Scenarios
Before using a GPU-accelerated ECS, make sure that the desired Tesla driver and CUDA toolkit have been installed on the ECS. Otherwise, computing acceleration will not take effect. This section describes how to obtain a Tesla driver and CUDA toolkit. Select a driver version based on your ECS type.
For instructions about how to install the Tesla driver and CUDA toolkit, see Manually Installing a Tesla Driver on a GPU-accelerated ECS.
Downloading a Tesla Driver
ECS Type |
Driver |
Product Series |
Product |
---|---|---|---|
P2s |
Tesla |
V |
V100 |
P2v |
Tesla |
V |
V100 |
Pi2 |
Tesla |
T |
T4 |
PI1 |
Tesla |
P |
P4 |
Downloading a CUDA Toolkit
Download the CUDA software package and select the corresponding CUDA Toolkit software package based on the instance type and driver version.
NVIDIA Driver Downloads provides the mapping between the driver version and CUDA Toolkit. If the versions do not match, the driver may be unavailable.
The following uses Tesla T4 as an example to describe how to download the driver package and CUDA Toolkit.
- Select the Linux operating system and the CUDA Toolkit 11.6 version.
Figure 1 Selecting the CUDA Toolkit version
- Select a CUDA Toolkit 11.6 package to download.
Figure 2 Downloading a CUDA Toolkit 11.6 package
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot