Comments (4)
Thanks! I tried a very high sampling rate (calling nvmlDeviceGetPowerUsage
without any delay) and the number of samples achieved in 1 second is around 2743, however, I don't think GPU power measurement hardware can have 2.7kHz sampling rate (I am using a Tesla K40m). I think the function returns the last measured power if we try to call it when the power measurement hardware is not ready or has recently returned a measured power from the sensor.
I will try the NVIDIA developer forums for exact sampling rate figures.
Closing this issue.
from gpustat.
We are now restricting the sampling rate to be lower than 10/sec (interval >= 0.1s), however you may try even higher rate to see what happens.
Line 127 in fb0f54e
from gpustat.
@europa1610
Could you please share a link to the forum post you refer to?
from gpustat.
Hey @piotr-dobrogost I posted my query on the NVIDIA Devtalk forums (NVML Subtopic) here but haven't received a response to my query yet. Here's my post for reference.
from gpustat.
Related Issues (20)
- --no-processes isn't present in the arguments, but is present in the readme HOT 1
- Some low-level errors (like `pynvml.nvml.NVMLError_LibRmVersionMismatch`) result in nothing printed (std or diagnostic) HOT 1
- UserWarning: Failed to setupterm(kind='xterm'): setupterm: could not find terminfo database HOT 1
- Support anaconda's legacy pynvml package HOT 7
- How to obtain RPM value for the fans ? HOT 2
- Plugin Architecture
- module 'pynvml' has no attribute '_nvmlGetFunctionPointer' HOT 17
- Truncate the "command" when use "-f" HOT 1
- ModuleNotFoundError: No module named '_curses' HOT 2
- ModuleNotFoundError: No module named '_curses' HOT 2
- Process not displayed HOT 3
- make appimage format or binary file οΌit can run everywhere HOT 1
- make appimage format HOT 1
- Error on querying NVIDIA devices | OverflowError: Python int too large to convert to C long HOT 9
- gpustat reports only the first program on nv driver 535 HOT 4
- Include GDDR6(X) VRAM temperatures HOT 2
- Show CUDA Driver Version in the output HOT 2
- Enhance gpustat to Display Latest CUDA Version Compatible with Current NVIDIA Driver HOT 4
- Even more compact (single-line) output for statusline use HOT 2
- Misreported used memory with the driver 535.129.03 HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
π Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. πππ
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google β€οΈ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from gpustat.