The benchmark tests were consistently performed utilizing the Frontier supercomputer at Oak Ridge National Laboratory (ORNL).
To initiate the benchmark execution, the primary step involves activating the data stream service. It's important to note that, for the purpose of testing, the data stream service has been installed on the login node of the Frontier system. The activation of this service can be achieved through the following command:
To initiate the benchmark execution, the primary step involves activating the data stream service. It is important to note that, for the purpose of testing, the data stream service has been installed on the login node of the Frontier system. The activation of this service can be achieved through the following command (preferably to be run on a separate shell):
```
source env.sh
python sender.py
python forwarder.py
```
The `job.sb` job description file provides the necessary steps for requesting computing nodes and setting the environment to run the benchmark.
@@ -46,6 +46,12 @@ The `job.sb` job description file provides the necessary steps for requesting co
sbatch job.sb
```
Once the job starts running, it will hold waiting for the data stream. To start streaming the data, execute the following command:
```
python sender.py
```
## Run Rules
- _Weak Scaling Experiments:_ Each rank at level 1 (refer to Figure 2) trains a TFT model on 64 ([4, 4, 4]) voxels, and the level 2 rank operates on [2, 2, 2] mean voxels of level 1. Consequently, for an input with dimensions 8 × 8 × 8, a total of 9 ranks (eight in level 1 and one in level 2) are required. For a larger input of 64 × 64 × 64, a total of 4608 ranks are needed, divided into 4096 in level 1 and 512 in level 2.
