Expected performance for 4x8H100 system

Hi! Many thanks for your contribution, I have question about performance.

test_internode.py results for WORLD_SIZE 4 version is [c18eabd]( https://github.com/deepseek-ai/DeepEP/tree/c18eabdebf1381978ff884d278f6083a6153be3f)
```
[1,3]:[tuning] SMs 24, NVL chunk 44, RDMA chunk 28, transmit: 6970.00 us, notify: 96.07 us, BW: 30.41 GB/s (RDMA), 60.77 GB/s (NVL)
[1,0]:[tuning] SMs 24, NVL chunk 44, RDMA chunk 32, transmit: 6919.00 us, notify: 152.74 us, BW: 30.63 GB/s (RDMA), 61.51 GB/s (NVL)
[1,0]:[tuning] Best dispatch (BF16): SMs 24, NVL chunk 40, RDMA chunk 4, transmit: 6881.00 us, notify: 160.81 us, BW: 30.80 GB/s (RDMA), 61.85 GB/s (NVL)
```

Does it perform as expected?

[full_log.txt](https://github.com/user-attachments/files/22231310/full_log.txt)

I also have warnings like this
```
WARN: ibgda_alloc_and_map_qp_uar with GPU as handler failed. We may need to enter the CPU fallback path.
WARN: cudaHostRegister with IoMemory failed with error=800. We may need to use a fallback path.
```

My setup:
```bash
> ibv_devinfo
hca_id: mlx5_0
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001a:677e
sys_image_guid: 58a2:e103:001a:677e
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3697
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_3
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001b:2416
sys_image_guid: 58a2:e103:001b:2416
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3705
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_4
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001a:2c6a
sys_image_guid: 58a2:e103:001a:2c6a
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3714
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_5
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001a:6766
sys_image_guid: 58a2:e103:001a:6766
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3713
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_6
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001a:2dfa
sys_image_guid: 58a2:e103:001a:2dfa
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3719
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_9
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001b:22c6
sys_image_guid: 58a2:e103:001b:22c6
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3732
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_10
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001b:242e
sys_image_guid: 58a2:e103:001b:242e
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3726
port_lmc: 0x00
link_layer: InfiniBand

hca_id: mlx5_11
transport: InfiniBand (0)
fw_ver: 28.41.1000
node_guid: 58a2:e103:001b:2136
sys_image_guid: 58a2:e103:001b:2136
vendor_id: 0x02c9
vendor_part_id: 4129
hw_ver: 0x0
board_id: DEL0000000036
phys_port_cnt: 1
port: 1
state: PORT_ACTIVE (4)
max_mtu: 4096 (5)
active_mtu: 4096 (5)
sm_lid: 5790
port_lid: 3735
port_lmc: 0x00
link_layer: InfiniBand
```

```bash
> nvidia-smi topo -mp
GPU0 GPU1 GPU2 GPU3 GPU4 GPU5 GPU6 GPU7 NIC0 NIC1 NIC2 NIC3 NIC4 NIC5 NIC6 NIC7 NIC8 NIC9 NIC10 NIC11 CPU Affinity NUMA Affinity GPU NUMA ID
GPU0 X NODE NODE NODE SYS SYS SYS SYS PIX PIX PIX NODE NODE NODE SYS SYS SYS SYS SYS SYS 0,2,4,6,8,10 0 N/A
GPU1 NODE X NODE NODE SYS SYS SYS SYS NODE NODE NODE PIX NODE NODE SYS SYS SYS SYS SYS SYS 0,2,4,6,8,10 0 N/A
GPU2 NODE NODE X NODE SYS SYS SYS SYS NODE NODE NODE NODE PIX NODE SYS SYS SYS SYS SYS SYS 0,2,4,6,8,10 0 N/A
GPU3 NODE NODE NODE X SYS SYS SYS SYS NODE NODE NODE NODE NODE PIX SYS SYS SYS SYS SYS SYS 0,2,4,6,8,10 0 N/A
GPU4 SYS SYS SYS SYS X NODE NODE NODE SYS SYS SYS SYS SYS SYS PIX PIX PIX NODE NODE NODE 1,3,5,7,9,11 1 N/A
GPU5 SYS SYS SYS SYS NODE X NODE NODE SYS SYS SYS SYS SYS SYS NODE NODE NODE PIX NODE NODE 1,3,5,7,9,11 1 N/A
GPU6 SYS SYS SYS SYS NODE NODE X NODE SYS SYS SYS SYS SYS SYS NODE NODE NODE NODE PIX NODE 1,3,5,7,9,11 1 N/A
GPU7 SYS SYS SYS SYS NODE NODE NODE X SYS SYS SYS SYS SYS SYS NODE NODE NODE NODE NODE PIX 1,3,5,7,9,11 1 N/A
NIC0 PIX NODE NODE NODE SYS SYS SYS SYS X PIX PIX NODE NODE NODE SYS SYS SYS SYS SYS SYS
NIC1 PIX NODE NODE NODE SYS SYS SYS SYS PIX X PIX NODE NODE NODE SYS SYS SYS SYS SYS SYS
NIC2 PIX NODE NODE NODE SYS SYS SYS SYS PIX PIX X NODE NODE NODE SYS SYS SYS SYS SYS SYS
NIC3 NODE PIX NODE NODE SYS SYS SYS SYS NODE NODE NODE X NODE NODE SYS SYS SYS SYS SYS SYS
NIC4 NODE NODE PIX NODE SYS SYS SYS SYS NODE NODE NODE NODE X NODE SYS SYS SYS SYS SYS SYS
NIC5 NODE NODE NODE PIX SYS SYS SYS SYS NODE NODE NODE NODE NODE X SYS SYS SYS SYS SYS SYS
NIC6 SYS SYS SYS SYS PIX NODE NODE NODE SYS SYS SYS SYS SYS SYS X PIX PIX NODE NODE NODE
NIC7 SYS SYS SYS SYS PIX NODE NODE NODE SYS SYS SYS SYS SYS SYS PIX X PIX NODE NODE NODE
NIC8 SYS SYS SYS SYS PIX NODE NODE NODE SYS SYS SYS SYS SYS SYS PIX PIX X NODE NODE NODE
NIC9 SYS SYS SYS SYS NODE PIX NODE NODE SYS SYS SYS SYS SYS SYS NODE NODE NODE X NODE NODE
NIC10 SYS SYS SYS SYS NODE NODE PIX NODE SYS SYS SYS SYS SYS SYS NODE NODE NODE NODE X NODE
NIC11 SYS SYS SYS SYS NODE NODE NODE PIX SYS SYS SYS SYS SYS SYS NODE NODE NODE NODE NODE X

Legend:

X = Self
SYS = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
PHB = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
PXB = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
PIX = Connection traversing at most a single PCIe bridge

NIC Legend:

NIC0: mlx5_0
NIC1: mlx5_1
NIC2: mlx5_2
NIC3: mlx5_3
NIC4: mlx5_4
NIC5: mlx5_5
NIC6: mlx5_6
NIC7: mlx5_7
NIC8: mlx5_8
NIC9: mlx5_9
NIC10: mlx5_10
NIC11: mlx5_11
```