# node 0
NCCL_IB_GID_INDEX=3 SUPPORT_CUTLASS_BLOCK_FP8=1 python -m sglang.launch_server \
--model-path /path/to/deepseek-ai__DeepSeek-R1 \
--tp-size 16 \
--dist-init-addr ${master_ip}:20001 \
--nnodes 2 \
--node-rank 0 \
--enable-ep-moe \
--enable-deepep-moe \
--deepep-mode auto \
--disable-cuda-graph \
--trust-remote-code \
--quantization fp8 \
--log-level info \
--chunked-prefill-size -1 \
--disable-radix-cache \
--context-length 65535 \
--max-running-requests 128 \
--stream-output \
> sglang-deep-ep-auto.log 2>&1 &

# node 1
NCCL_IB_GID_INDEX=3 SUPPORT_CUTLASS_BLOCK_FP8=1 python -m sglang.launch_server \
--model-path /path/to/deepseek-ai__DeepSeek-R1 \
--tp-size 16 \
--dist-init-addr ${master_ip}:20001 \
--nnodes 2 \
--node-rank 1 \
--enable-ep-moe \
--enable-deepep-moe \
--deepep-mode auto \
--disable-cuda-graph \
--trust-remote-code \
--quantization fp8 \
--log-level info \
--chunked-prefill-size -1 \
--disable-radix-cache \
--context-length 65535 \
--max-running-requests 128 \
--stream-output \
> sglang-deep-ep-auto.log 2>&1 &

$ python3.11 -m sglang.check_env

Python: 3.11.11 (main, Dec 16 2024, 17:23:09) [GCC 10.2.1 20200825 (Alibaba 10.2.1-3.8 2.32)]
CUDA available: True
GPU 0,1,2,3,4,5,6,7: NVIDIA H20
GPU 0,1,2,3,4,5,6,7 Compute Capability: 9.0
CUDA_HOME: /usr/local/cuda
NVCC: Cuda compilation tools, release 12.4, V12.4.99
CUDA Driver Version: 550.127.08
PyTorch: 2.6.0+cu124
sglang: 0.4.5
sgl_kernel: 0.0.6
flashinfer: Module Not Found
triton: 3.2.0
transformers: 4.50.3
torchao: 0.9.0
numpy: 1.26.4
aiohttp: 3.11.16
fastapi: 0.115.12
hf_transfer: Module Not Found
huggingface_hub: 0.30.1
interegular: 0.3.3
modelscope: Module Not Found
orjson: 3.10.16
outlines: 0.1.11
packaging: 24.2
psutil: 7.0.0
pydantic: 2.11.1
multipart: Module Not Found
zmq: 0.0.0
uvicorn: 0.34.0
uvloop: 0.21.0
vllm: 0.8.2
xgrammar: 0.1.16
openai: 1.70.0
tiktoken: 0.9.0
anthropic: Module Not Found
litellm: Module Not Found
decord: 0.6.0
NVIDIA Topology: 
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    NIC0    NIC1    NIC2    NIC3    CPU Affinity    NUMA Affinity  GPU NUMA ID
GPU0     X      NV18    NV18    NV18    NV18    NV18    NV18    NV18    NODE    NODE    SYS     SYS     0-47,96-143     0     N/A
GPU1    NV18     X      NV18    NV18    NV18    NV18    NV18    NV18    PIX     NODE    SYS     SYS     0-47,96-143     0     N/A
GPU2    NV18    NV18     X      NV18    NV18    NV18    NV18    NV18    NODE    NODE    SYS     SYS     0-47,96-143     0     N/A
GPU3    NV18    NV18    NV18     X      NV18    NV18    NV18    NV18    NODE    PIX     SYS     SYS     0-47,96-143     0     N/A
GPU4    NV18    NV18    NV18    NV18     X      NV18    NV18    NV18    SYS     SYS     PIX     NODE    48-95,144-191   1     N/A
GPU5    NV18    NV18    NV18    NV18    NV18     X      NV18    NV18    SYS     SYS     NODE    NODE    48-95,144-191   1     N/A
GPU6    NV18    NV18    NV18    NV18    NV18    NV18     X      NV18    SYS     SYS     NODE    PIX     48-95,144-191   1     N/A
GPU7    NV18    NV18    NV18    NV18    NV18    NV18    NV18     X      SYS     SYS     NODE    NODE    48-95,144-191   1     N/A
NIC0    NODE    PIX     NODE    NODE    SYS     SYS     SYS     SYS      X      NODE    SYS     SYS
NIC1    NODE    NODE    NODE    PIX     SYS     SYS     SYS     SYS     NODE     X      SYS     SYS
NIC2    SYS     SYS     SYS     SYS     PIX     NODE    NODE    NODE    SYS     SYS      X      NODE
NIC3    SYS     SYS     SYS     SYS     NODE    NODE    PIX     NODE    SYS     SYS     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
  NV#  = Connection traversing a bonded set of # NVLinks

NIC Legend:

  NIC0: mlx5_bond_0
  NIC1: mlx5_bond_1
  NIC2: mlx5_bond_2
  NIC3: mlx5_bond_3


ulimit soft: 204800

export NCCL_IB_DISABLE=0
export NCCL_SOCKET_IFNAME=bond0
export NCCL_NET_PLUGIN=''
export NCCL_IB_GID_INDEX=3
export NCCL_IB_TIMEOUT=22
export NCCL_IB_RETRY_CNT=7
export NCCL_IB_SL=5
export NCCL_IB_TC=136
export NCCL_IB_HCA="mlx5_bond_0,mlx5_bond_1,mlx5_bond_2,mlx5_bond_3"
export NCCL_DEBUG=WARN
export NCCL_SET_THREAD_NAME=1
export NCCL_IB_QPS_PER_CONNECTION=8
export NCCL_SET_THREAD_NAME=1
export NCCL_DEBUG_SUBSYS=INIT,TUNING,GRAPH

export NVSHMEM_DEBUG=WARN
export NVSHMEM_ENABLE_NIC_PE_MAPPING=1
export NVSHMEM_HCA_PE_MAPPING="mlx5_bond_0:1:2,mlx5_bond_1:1:2,mlx5_bond_2:1:2,mlx5_bond_3:1:2"
export NVSHMEM_IBRC_ROCE_LAG_PORT_SELECTION=3
export NVSHMEM_IB_GID_INDEX=3
export NVSHMEM_IB_TRAFFIC_CLASS=64

export GLOO_SOCKET_IFNAME=eth0

$ LD_LIBRARY_PATH=/opt/gdrcopy/lib /opt/gdrcopy/bin/gdrcopy_copybw
GPU id:0; name: NVIDIA H20; Bus id: 0000:08:00
GPU id:1; name: NVIDIA H20; Bus id: 0000:7e:00
GPU id:2; name: NVIDIA H20; Bus id: 0000:a2:00
GPU id:3; name: NVIDIA H20; Bus id: 0000:c6:00
GPU id:4; name: NVIDIA H20; Bus id: 0001:09:00
GPU id:5; name: NVIDIA H20; Bus id: 0001:7f:00
GPU id:6; name: NVIDIA H20; Bus id: 0001:a3:00
GPU id:7; name: NVIDIA H20; Bus id: 0001:c7:00
selecting device 0
testing size: 131072
rounded size: 131072
gpu alloc fn: cuMemAlloc
device ptr: 7f825d400000
map_d_ptr: 0x7f848c0a6000
info.va: 7f825d400000
info.mapped_size: 131072
info.page_size: 65536
info.mapped: 1
info.wc_mapping: 1
page offset: 0
user-space pointer:0x7f848c0a6000
writing test, size=131072 offset=0 num_iters=10000
write BW: 17993.9MB/s
reading test, size=131072 offset=0 num_iters=100
read BW: 716.355MB/s
unmapping buffer
unpinning buffer
closing gdrdrv

$ LD_LIBRARY_PATH=/opt/nvshmem/lib/ /opt/nvshmem/bin/nvshmem-info -a
NVSHMEM v3.2.5

Build Information:
  CUDA API                     12040
  CUDA Driver                  12040
  Build Timestamp              Apr  2 2025 14:47:31
  Build Variables             
        NVSHMEM_DEBUG=OFF NVSHMEM_DEVEL=OFF NVSHMEM_DEFAULT_PMI2=OFF
        NVSHMEM_DEFAULT_PMIX=OFF NVSHMEM_DEFAULT_UCX=OFF
        NVSHMEM_ENABLE_ALL_DEVICE_INLINING=OFF NVSHMEM_GPU_COLL_USE_LDST=OFF
        NVSHMEM_IBGDA_SUPPORT=ON NVSHMEM_IBGDA_SUPPORT_GPUMEM_ONLY=OFF
        NVSHMEM_IBDEVX_SUPPORT=OFF NVSHMEM_IBRC_SUPPORT=ON
        NVSHMEM_MPI_SUPPORT=OFF NVSHMEM_NVTX=ON NVSHMEM_PMIX_SUPPORT=OFF
        NVSHMEM_SHMEM_SUPPORT=OFF NVSHMEM_TEST_STATIC_LIB=OFF
        NVSHMEM_TIMEOUT_DEVICE_POLLING=OFF NVSHMEM_TRACE=OFF NVSHMEM_UCX_SUPPORT=OFF
        NVSHMEM_USE_DLMALLOC=OFF NVSHMEM_USE_NCCL=OFF NVSHMEM_USE_GDRCOPY=ON
        NVSHMEM_VERBOSE=OFF CUDA_HOME=/usr/local/cuda GDRCOPY_HOME=/opt/gdrcopy
        LIBFABRIC_HOME=/usr/local/libfabric MPI_HOME=/usr/local/ompi
        NCCL_HOME=/usr/local/nccl NVSHMEM_PREFIX=/usr/local/nvshmem PMIX_HOME=/usr
        SHMEM_HOME=/usr/local/ompi UCX_HOME=/usr/local/ucx

Standard options:
  NVSHMEM_VERSION              false (type: bool, default: false)
        Print library version at startup
  NVSHMEM_INFO                 false (type: bool, default: false)
        Print environment variable options at startup
  NVSHMEM_DISABLE_NVLS         false (type: bool, default: false)
        Disable NVLS SHARP resources for collectives, even if available for platform
  NVSHMEM_SYMMETRIC_SIZE       1073741824 (type: size, default: 1073741824)
        Specifies the size (in bytes) of the symmetric heap memory per PE. The
        size is implementation-defined and must be at least as large as the integer
        ceiling of the product of the numeric prefix and the scaling factor. The
        character suffixes for the scaling factor are as follows:

          *  k or K multiplies by 2^10 (kibibytes)
          *  m or M multiplies by 2^20 (mebibytes)
          *  g or G multiplies by 2^30 (gibibytes)
          *  t or T multiplies by 2^40 (tebibytes)

        For example, string '20m' is equivalent to the integer value 20971520, or 20
        mebibytes. Similarly the string '3.1M' is equivalent to the integer value
        3250586. Only one multiplier is recognized and any characters following the
        multiplier are ignored, so '20kk' will not produce the same result as '20m'.
        Usage of string '.5m' will yield the same result as the string '0.5m'.
        An invalid value for NVSHMEM_SYMMETRIC_SIZE is an error, which the NVSHMEM
        library shall report by either returning a nonzero value from
        nvshmem_init_thread or causing program termination.
  NVSHMEM_DEBUG                "WARN" (type: string, default: "")
        Set to enable debugging messages.
        Optional values: VERSION, WARN, INFO, ABORT, TRACE

Bootstrap options:
  NVSHMEM_BOOTSTRAP            "PMI" (type: string, default: "PMI")
        Name of the default bootstrap that should be used to initialize NVSHMEM.
        Allowed values: PMI, MPI, SHMEM, plugin, UID
  NVSHMEM_BOOTSTRAP_PMI        "PMI" (type: string, default: "PMI")
        Name of the PMI bootstrap that should be used to initialize NVSHMEM.
        Allowed values: PMI, PMI-2, PMIX
  NVSHMEM_BOOTSTRAP_PLUGIN     "" (type: string, default: "")
        Absolute path to or name of the bootstrap plugin file to load when
        NVSHMEM_BOOTSTRAP=plugin is specified
  NVSHMEM_BOOTSTRAP_MPI_PLUGIN "nvshmem_bootstrap_mpi.so.3" (type: string, default: "nvshmem_bootstrap_mpi.so.3")
        Absolute path to or name of the MPI bootstrap plugin file. 
        NVSHMEM will search for the plugin based on linux linker priorities. See man
        dlopen
  NVSHMEM_BOOTSTRAP_SHMEM_PLUGIN "nvshmem_bootstrap_shmem.so.3" (type: string, default: "nvshmem_bootstrap_shmem.so.3")
        Absolute path to or name of the SHMEM bootstrap plugin file. 
        NVSHMEM will search for the plugin based on linux linker priorities. See man
        dlopen
  NVSHMEM_BOOTSTRAP_PMI_PLUGIN "nvshmem_bootstrap_pmi.so.3" (type: string, default: "nvshmem_bootstrap_pmi.so.3")
        Absolute path to or name of the PMI bootstrap plugin file. 
        NVSHMEM will search for the plugin based on linux linker priorities. See man
        dlopen
  NVSHMEM_BOOTSTRAP_PMI2_PLUGIN "nvshmem_bootstrap_pmi2.so.3" (type: string, default: "nvshmem_bootstrap_pmi2.so.3")
        Absolute path to or name of the PMI-2 bootstrap plugin file. 
        NVSHMEM will search for the plugin based on linux linker priorities. See man
        dlopen
  NVSHMEM_BOOTSTRAP_PMIX_PLUGIN "nvshmem_bootstrap_pmix.so.3" (type: string, default: "nvshmem_bootstrap_pmix.so.3")
        Absolute path to or name of the PMIx bootstrap plugin file. 
        NVSHMEM will search for the plugin based on linux linker priorities. See man
        dlopen
  NVSHMEM_BOOTSTRAP_UID_PLUGIN "nvshmem_bootstrap_uid.so.3" (type: string, default: "nvshmem_bootstrap_uid.so.3")
        Absolute path to or name of the UID bootstrap plugin file. 
        NVSHMEM will search for the plugin based on linux linker priorities. See man
        dlopen

Additional options:
  NVSHMEM_CUDA_PATH            "" (type: string, default: "")
        Path to directory containing libcuda.so (for use when not in default location)
  NVSHMEM_DEBUG_ATTACH_DELAY   0 (type: int, default: 0)
        Delay (in seconds) during the first call to NVSHMEM_INIT to allow for attaching
        a debuggger (Default 0)
  NVSHMEM_DEBUG_FILE           "" (type: string, default: "")
        Debugging output filename, may contain %h for hostname and %p for pid
  NVSHMEM_MAX_TEAMS            32 (type: long, default: 32)
        Maximum number of simultaneous teams allowed
  NVSHMEM_MAX_MEMORY_PER_GPU   137438953472 (type: size, default: 137438953472)
        Maximum memory per GPU
  NVSHMEM_DISABLE_CUDA_VMM     false (type: bool, default: false)
        Disable use of CUDA VMM for P2P memory mapping. By default, CUDA VMM is enabled
        on x86 and disabled on P9. CUDA VMM feature in NVSHMEM requires CUDA RT version
        and CUDA Driver version to be greater than or equal to 11.3.
  NVSHMEM_DISABLE_P2P          false (type: bool, default: false)
        Disable P2P connectivity of GPUs even when available
  NVSHMEM_IGNORE_CUDA_MPS_ACTIVE_THREAD_PERCENTAGE false (type: bool, default: false)
        When doing Multi-Process Per GPU (MPG) run, full API support is available only
        if sum of CUDA_MPS_ACTIVE_THREAD_PERCENTAGE of processes running on a GPU is <=
        100%. Through this variable, user can request NVSHMEM runtime to ignore the
        active thread percentage and allow full MPG support. Users enable it at their
        own risk as NVSHMEM might deadlock.
  NVSHMEM_CUMEM_GRANULARITY    536870912 (type: size, default: 536870912)
        Granularity for cuMemAlloc/cuMemCreate
  NVSHMEM_PROXY_REQUEST_BATCH_MAX 32 (type: int, default: 32)
        Maxmum number of requests that the proxy thread processes in a single iteration
        of the progress loop.

Collectives options:
  NVSHMEM_DISABLE_NCCL         false (type: bool, default: false)
        Disable use of NCCL for collective operations
  NVSHMEM_BARRIER_DISSEM_KVAL  2 (type: int, default: 2)
        Radix of the dissemination algorithm used for barriers
  NVSHMEM_BARRIER_TG_DISSEM_KVAL 2 (type: int, default: 2)
        Radix of the dissemination algorithm used for thread group barriers
  NVSHMEM_FCOLLECT_LL_THRESHOLD 2048 (type: size, default: 2048)
        Message size threshold up to which fcollect LL algo will be used

  NVSHMEM_REDUCE_SCRATCH_SIZE  524288 (type: size, default: 524288)
        Amount of symmetric heap memory (minimum 16B, multiple of 8B) reserved by
        runtime for every team to implement reduce and reducescatter collectives

  NVSHMEM_BCAST_ALGO           0 (type: int, default: 0)
        Broadcast algorithm to be used.
          * 0 - use default algorithm selection strategy

  NVSHMEM_REDMAXLOC_ALGO       1 (type: int, default: 1)
        Reduction algorithm to be used for MAXLOC operation.
          * 1 - default, flag alltoall algorithm
          * 2 - flat reduce + flat bcast
          * 3 - topo-aware two-level reduce + topo-aware bcast


Transport options:
  NVSHMEM_REMOTE_TRANSPORT     "ibrc" (type: string, default: "ibrc")
        Selected transport for remote operations: ibrc, ucx, libfabric, ibdevx, none
  NVSHMEM_ENABLE_NIC_PE_MAPPING true (type: bool, default: false)
        When not set or set to 0, a PE is assigned the NIC on the node that is closest
        to it by distance. When set to 1, NVSHMEM either assigns NICs to PEs on a
        round-robin basis or uses NVSHMEM_HCA_PE_MAPPING or NVSHMEM_HCA_LIST when they
        are specified.
  NVSHMEM_DISABLE_LOCAL_ONLY_PROXY false (type: bool, default: false)
        When running on an NVLink-only configuaration (No-IB, No-UCX), completely
        disable the proxy thread. This will disable device side global exit and device
        side wait timeout polling (enabled by NVSHMEM_TIMEOUT_DEVICE_POLLING build-time
        variable) because these are processed by the proxy thread.
  NVSHMEM_IB_ENABLE_IBGDA      false (type: bool, default: false)
        Set to enable GPU-initiated communication transport.

Hidden options:
  NVSHMEM_INFO_HIDDEN          true (type: bool, default: false)
        Print hidden environment variable options at startup
  NVSHMEM_DISABLE_NVLS_SHARING true (type: bool, default: true)
        Disable NVLS SHARP resource sharing for user-defined teams
  NVSHMEM_HEAP_KIND            "DEVICE" (type: string, default: "DEVICE")
        Specify the memory kind used by the NVSHMEM symmetric heap.
        Allowed values: VIDMEM, SYSMEM
  NVSHMEM_ENABLE_RAIL_OPT      false (type: bool, default: false)
        Enable Rail Optimization when heap is in SYSMEM
  NVSHMEM_BOOTSTRAP_TWO_STAGE  false (type: bool, default: false)
        Ignore CUDA device setting during initialization,forcing two-stage
        initialization
  NVSHMEM_DEBUG_SUBSYS         "" (type: string, default: "")
        Comma separated list of debugging message sources. Prefix with '^' to exclude.
        Values: INIT, COLL, P2P, PROXY, TRANSPORT, MEM, BOOTSTRAP, TOPO, UTIL, ALL
  NVSHMEM_ENABLE_ERROR_CHECKS  false (type: bool, default: false)
        Enable error checks
  NVSHMEM_DISABLE_MNNVL        false (type: bool, default: false)
        Disable MNNVL connectivity for GPUs even when available
  NVSHMEM_CUMEM_HANDLE_TYPE    "FILE_DESCRIPTOR" (type: string, default: "FILE_DESCRIPTOR")
        Handle type for cuMemCreate. Supported are - FABRIC or FILE_DESCRIPTOR
  NVSHMEM_BYPASS_ACCESSIBILITY_CHECK false (type: bool, default: false)
        Bypass peer GPU accessbility checks
  NVSHMEM_FCOLLECT_NTHREADS    512 (type: int, default: 512)
        Sets number of threads per block for fcollect collective.
        By default, if no env is set, default value is min(max_occupancy per CTA, msg
        size per PE).
        If env is specified, value overrides the default irrespective of max occupancy
        per CTA

  NVSHMEM_REDUCESCATTER_NTHREADS 512 (type: int, default: 512)
        Sets number of threads per block for reducescatter collective.
        By default, if no env is set, default value is min(max_occupancy per CTA, msg
        size per PE).
        If env is specified, value overrides the default irrespective of max occupancy
        per CTA

  NVSHMEM_MAX_CTAS             1 (type: int, default: 1)
        Sets number of blocks per grid for host onstream collective.
        By default, if no env is set, default value to 1 CTA
        If env is specified, value overrides the default value

  NVSHMEM_REDUCE_RECEXCH_KVAL  2 (type: int, default: 2)
        Radix of the recursive exchange reduction algorithm
  NVSHMEM_FCOLLECT_LL128_THRESHOLD 0 (type: size, default: 0)
        Message size threshold up to which the fcollect LL128 algo will be used.
        LL128 will be used only when FCOLLECT_LL_THRESHOLD < size
  NVSHMEM_FCOLLECT_NVLS_THRESHOLD 16777216 (type: size, default: 16777216)
        Message size threshold up to which fcollect NVLS algo will be used

  NVSHMEM_REDUCESCATTER_NVLS_THRESHOLD 16777216 (type: size, default: 16777216)
        Message size threshold up to which reducescatter NVLS algo will be used

  NVSHMEM_BCAST_TREE_KVAL      2 (type: int, default: 2)
        Radix of the broadcast tree algorithm
  NVSHMEM_FCOLLECT_ALGO        0 (type: int, default: 0)
        Fcollect algorithm to be used. 
          * 0 - use default algorithm selection strategy

  NVSHMEM_REDUCE_ALGO          0 (type: int, default: 0)
        Allreduce algorithm to be used. 
           * 0/1 - use default algorithm selection strategy

  NVSHMEM_REDUCE_NVLS_THRESHOLD 2048 (type: int, default: 2048)
        Message size threshold up to which allreduce one-shot algo will be used

  NVSHMEM_REDUCESCATTER_ALGO   0 (type: int, default: 0)
        Reduce Scatter algorithm to be used. 
          * 0 - use default algorithm selection strategy

  NVSHMEM_ASSERT_ATOMICS_SYNC  false (type: bool, default: false)
        Bypass flush on wait_until at target
  NVSHMEM_BYPASS_FLUSH         false (type: bool, default: false)
        Bypass flush in proxy when enforcing consistency

NVTX options:
  NVSHMEM_NVTX                 "off" (type: string, default: "off")
        Set to enable NVTX instrumentation. Accepts a comma separated list of
        instrumentation groups. By default the NVTX instrumentation is disabled.
          init                : library setup
          alloc               : memory management
          launch              : kernel launch routines
          coll                : collective communications
          wait                : blocking point-to-point synchronization
          wait_on_stream      : point-to-point synchronization (on stream)
          test                : non-blocking point-to-point synchronization
          memorder            : memory ordering (quiet, fence)
          quiet_on_stream     : nvshmemx_quiet_on_stream
          atomic_fetch        : fetching atomic memory operations
          atomic_set          : non-fetchong atomic memory operations
          rma_blocking        : blocking remote memory access operations
          rma_nonblocking     : non-blocking remote memory access operations
          proxy               : activity of the proxy thread
          common              : init,alloc,launch,coll,memorder,wait,atomic_fetch,rma_blocking,proxy
          all                 : all groups
          off                 : disable all NVTX instrumentation