Commit 2dd2436e authored by Ferreira Da Silva, Rafael's avatar Ferreira Da Silva, Rafael
Browse files

Update README.md

parent 472770ef
Loading
Loading
Loading
Loading
+7 −7
Original line number Diff line number Diff line
@@ -61,7 +61,7 @@ python sender.py

## Figure of Merit

The primary figure of merit for the ML4NSE (defined in detail in https://doi.org/10.1615/JMachLearnModelComput.2023048607) workflow benchmark is time-to-solution. Some useful secondary figures of merit will be:
The primary figure of merit for the ML4NSE (defined in detail in https://doi.org/10.1615/JMachLearnModelComput.2023048607) workflow benchmark is computation throughput (i.e., the inverse of time-to-solution). Some useful secondary figures of merit will be:

- Exchange bandwidth (ingress) at the gateway node which multiplexes among input streams

@@ -69,10 +69,10 @@ The primary figure of merit for the ML4NSE (defined in detail in https://doi.org

An additional potential figure of merit that could demonstrate the robustness of the system would include any active guidance between the Compute and Services Clusters; the latency involved in control operations becomes crucial. Specifically, it's essential to assess the duration a compute job is held while disseminating new control information. Also, as additional input streams and output consumers are added, the effect on end-to-end time-to-solution could be affected.

| Dataset | Dimension | # Nodes | Sending Transfer Rate | Avg. Receiving Transfer Rate (per rank) | Execution Time |
| Dataset | Dimension | # Nodes | Sending Transfer Rate | Avg. Receiving Transfer Rate (per rank) | Throughput |
| ------ | ------ | ------ | ------ | ------ | ------ |
| p_322_data_np_res_16.npy | 16 x 16 x 16 | 9 | 62.79 Gbps | 10.43 Mbps | 874.03s |
| p_322_data_np_res_32.npy | 16 x 16 x 16 | 9 | 68.68 Gbps | 125.72 Mbps | 883.70s |
| p_322_data_np_res_64.npy | 16 x 16 x 16 | 9 | 62.17 Gbps | 555.57 Mbps | 893.30s |
| p_322_data_np_res_32.npy | 32 x 32 x 32 | 72 | 67.17 Gbps | 60.26 Mbps | 1077.95s |
| p_322_data_np_res_64.npy | 32 x 32 x 32 | 72 | 70.08 Gbps | 95.51 Mbps | 1571.26s |
| p_322_data_np_res_16.npy | 16 x 16 x 16 | 9 | 62.79 Gbps | 10.43 Mbps | 1.144 × 10<sup>-3</sup> |
| p_322_data_np_res_32.npy | 16 x 16 x 16 | 9 | 68.68 Gbps | 125.72 Mbps | 1.131 × 10<sup>-3</sup> |
| p_322_data_np_res_64.npy | 16 x 16 x 16 | 9 | 62.17 Gbps | 555.57 Mbps | 1.112 × 10<sup>-3</sup> |
| p_322_data_np_res_32.npy | 32 x 32 x 32 | 72 | 67.17 Gbps | 60.26 Mbps | 0.927 × 10<sup>-3</sup> |
| p_322_data_np_res_64.npy | 32 x 32 x 32 | 72 | 70.08 Gbps | 95.51 Mbps | 0.636 × 10<sup>-3</sup> |