c++ tool to perform NVIDIA Fabric Manager operations, extending the functionalities from NVIDIA sample code.
root@localhost:~/NVIDIA-FabricManager-partition-tool# ./partitioner -h
Usage: ./partitioner [-i <IP>] -o <OP> [-p <ID>]
-i, --ip <IP> : IP address of Fabric Manager (default: 127.0.0.1)
-o, --operation <N>: 0=List, 1=Activate, 2=Deactivate
-p, --partition <ID>: Partition ID (required for Activate/Deactivate)
Running without options starts interactive mode.
root@localhost:~/NVIDIA-FabricManager-partition-tool# ./partitioner -o 0
Successfully connected to Fabric Manager at 127.0.0.1
Total supported partitions: 15
Partition ID Number of GPUs GPU Module ID Max NVLinks/GPU STATUS
--------------------------------------------------------------------------------
0 8 1, 2, 3, 4, 5, 6, 7, 8 18 INACTIVE
1 4 1, 2, 3, 4 18 INACTIVE
2 4 5, 6, 7, 8 18 INACTIVE
3 2 1, 3 18 INACTIVE
4 2 2, 4 18 INACTIVE
5 2 5, 7 18 INACTIVE
6 2 6, 8 18 ACTIVE
7 1 1 0 INACTIVE
8 1 2 0 INACTIVE
9 1 3 0 INACTIVE
10 1 4 0 INACTIVE
11 1 5 0 INACTIVE
12 1 6 0 INACTIVE
13 1 7 0 INACTIVE
14 1 8 0 INACTIVE
root@localhost:~/NVIDIA-FabricManager-partition-tool# ./partitioner -o 0 -f csv
Partition_ID,Num_GPUs,GPU_Module_IDs,Max_NVLinks_per_GPU,Status
0,8,"1 2 3 4 5 6 7 8",18,INACTIVE
1,4,"1 2 3 4",18,INACTIVE
2,4,"5 6 7 8",18,INACTIVE
3,2,"1 3",18,INACTIVE
4,2,"2 4",18,INACTIVE
5,2,"5 7",18,INACTIVE
6,2,"6 8",18,ACTIVE
7,1,"1",0,INACTIVE
8,1,"2",0,INACTIVE
9,1,"3",0,INACTIVE
10,1,"4",0,INACTIVE
11,1,"5",0,INACTIVE
12,1,"6",0,INACTIVE
13,1,"7",0,INACTIVE
14,1,"8",0,INACTIVE
root@localhost:~/NVIDIA-FabricManager-partition-tool# ./partitioner -o 2 -p 5
Connecting to Fabric Manager at 127.0.0.1...
Successfully connected to Fabric Manager at 127.0.0.1
Successfully sent deactivation request for partition 5
