67 lines
2.5 KiB
Bash
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# TP×DP matrix experiment for DeepSeek-V4-Flash on H200 (8 GPUs) using SGLang Docker.
# Tests SGLang with three parallel configurations:
# TP=2, DP=4 -> 2 GPUs per replica, 4 replicas
# TP=4, DP=2 -> 4 GPUs per replica, 2 replicas
# TP=8, DP=1 -> 8 GPUs, no data parallelism
EXPERIMENT="dsv4_h200_sglang_tp_dp_matrix"
MODEL_NAME="DeepSeek-V4-Flash"
MODEL_PATH="/data3/hf_models/DeepSeek-V4-Flash"
SERVED_MODEL_NAME="deepseek-v4-flash"
SGLANG_PORT="${SGLANG_PORT:-30031}"
# Python interpreter for orchestration scripts (parse_backend.py, compare.py, etc.)
# and the benchmark client. Defaults to the system python3 if the sglang venv
# does not exist on the host.
VENV_CLIENT="${VENV_CLIENT:-/data/user1/yy/envs/sglang}"
# When USE_DOCKER_CLIENT=1, the benchmark client (sglang.bench_serving) is run
# inside the same SGLang Docker image used for the server. This avoids requiring
# a native sglang installation on the host.
USE_DOCKER_CLIENT="${USE_DOCKER_CLIENT:-1}"
export CUDA_VISIBLE_DEVICES="${CUDA_VISIBLE_DEVICES:-0,1,2,3,4,5,6,7}"
# Runtime working directory for logs, pid files, and tmp. Defaults to a local
# directory under this experiment so the benchmark is self-contained.
RUNTIME_BASE="${RUNTIME_BASE:-${SCRIPT_DIR}/runtime}"
# Parallel configurations to test. Format: "TP DP"
declare -a PARALLEL_CONFIGS=(
"2 4"
"4 2"
"8 1"
)
# SGLang server settings
CONTEXT_LENGTH="${CONTEXT_LENGTH:-65536}"
MAX_RUNNING_REQUESTS="${MAX_RUNNING_REQUESTS:-64}"
MEM_FRACTION_STATIC="${MEM_FRACTION_STATIC:-0.88}"
MOE_RUNNER_BACKEND="${MOE_RUNNER_BACKEND:-marlin}"
# Docker deployment switch. SGLang is launched via Docker using the image below.
USE_DOCKER="${USE_DOCKER:-1}"
DOCKER_IMAGE="${DOCKER_IMAGE:-lmsysorg/sglang:latest}"
# Matrix and concurrency rules are defined in matrix.json by default.
MATRIX_FILE="${MATRIX_FILE:-${SCRIPT_DIR:-.}/matrix.json}"
MATRIX_MODE="${MATRIX_MODE:-Y}"
# Sampling density for concurrency. 0 means use the default heuristic in
# generate_scenarios.py (6-8 points, or all integers when range is small).
CONCURRENCY_SAMPLES="${CONCURRENCY_SAMPLES:-0}"
# Per-scenario timeout to avoid hangs (seconds).
SCENARIO_TIMEOUT_S="${SCENARIO_TIMEOUT_S:-1800}"
# GPU memory sampling interval (seconds).
GPU_MEM_SAMPLE_INTERVAL_S="${GPU_MEM_SAMPLE_INTERVAL_S:-1}"
# Dry-run mode: if 1, only log the server args and scenario plan without starting
# any server or sending requests.
DRY_RUN="${DRY_RUN:-0}"
# Per-config scenario limit for quick smoke tests. 0 = run all generated scenarios.
GRID_LIMIT="${GRID_LIMIT:-0}"