Gahow Wang
547611e022
D2: run_benchmark.sh and run_experiments.sh still pass --time-scale and --max-inflight-sessions to the replayer, but those flags were removed when the project moved to trace-driven dispatch. The scripts cannot run as-is. D3: ~25 ad-hoc analyze_* / compare_* / profile_* / final_* scripts and a handful of single-experiment run_*.sh point at /home/admin/cpfs paths, deleted output directories, or a sampled trace file that no longer exists. Keep them in scripts/legacy/ for historical reference; the scripts that remain in scripts/ (analyze_trace, analyze_breakdown, analyze_cache_hit, analyze_eviction, compare_results, compute_roofline, sample_trace, analyze_agentic_patterns, simulate_cache_policies, plus launch_*.sh, gpu_monitor.sh, bench.sh) cover the current workflow. Adds scripts/legacy/README.md to document the archival policy.
101 lines
4.4 KiB
Bash
Executable File
101 lines
4.4 KiB
Bash
Executable File
#!/bin/bash
|
|
# PS + flexible D: HEAVY prefill on PS, decode on least-loaded C
|
|
set -euo pipefail
|
|
cd /home/admin/cpfs/wjh/agentic-kv
|
|
source .venv/bin/activate
|
|
|
|
MODEL=/home/admin/cpfs/wjh/models/Qwen/Qwen3-Coder-30B-A3B-Instruct
|
|
OUTDIR=outputs/ps_flexd
|
|
|
|
cleanup() {
|
|
for p in $(ps aux | grep -E 'vllm serve|cache_aware_proxy' | grep -v grep | awk '{print $2}' 2>/dev/null); do kill -9 "$p" 2>/dev/null || true; done
|
|
sleep 3
|
|
for p in $(fuser /dev/nvidia* 2>/dev/null | tr ' ' '\n' | sort -u | grep -v '^$' || true); do kill -9 "$p" 2>/dev/null || true; done
|
|
sleep 5
|
|
}
|
|
|
|
cleanup
|
|
mkdir -p "$OUTDIR"
|
|
|
|
# 7 C instances (kv_both, GPUs 0-6)
|
|
for i in $(seq 0 6); do
|
|
VLLM_MOONCAKE_BOOTSTRAP_PORT=$((8998+i)) MASTER_PORT=$((29500+i)) CUDA_VISIBLE_DEVICES=$i \
|
|
.venv/bin/vllm serve "$MODEL" --host 0.0.0.0 --port $((8000+i)) --tensor-parallel-size 1 \
|
|
--trust-remote-code --enable-prefix-caching --enforce-eager \
|
|
--dtype auto --gpu-memory-utilization 0.9 --max-model-len 200000 \
|
|
--kv-transfer-config '{"kv_connector":"MooncakeConnector","kv_role":"kv_both"}' \
|
|
> "$OUTDIR/vllm_c_$i.log" 2>&1 &
|
|
sleep 2
|
|
done
|
|
|
|
# 1 PS instance (kv_both, GPU 7)
|
|
VLLM_MOONCAKE_BOOTSTRAP_PORT=9005 MASTER_PORT=29507 CUDA_VISIBLE_DEVICES=7 \
|
|
.venv/bin/vllm serve "$MODEL" --host 0.0.0.0 --port 8007 --tensor-parallel-size 1 \
|
|
--trust-remote-code --enable-prefix-caching --enforce-eager \
|
|
--dtype auto --gpu-memory-utilization 0.9 --max-model-len 200000 \
|
|
--kv-transfer-config '{"kv_connector":"MooncakeConnector","kv_role":"kv_both"}' \
|
|
> "$OUTDIR/vllm_ps.log" 2>&1 &
|
|
|
|
echo "Waiting for instances..."
|
|
for i in $(seq 0 7); do
|
|
timeout 600 bash -c "until curl -s localhost:$((8000+i))/health > /dev/null 2>&1; do sleep 5; done"
|
|
echo " inst_$i healthy"
|
|
done
|
|
for bp in $(seq 8998 9005); do
|
|
timeout 120 bash -c "until curl -s localhost:$bp/query > /dev/null 2>&1; do sleep 2; done"
|
|
done
|
|
echo "All ready"
|
|
|
|
# Proxy: PS on port 8007, C on 8000-8006, flexible D
|
|
.venv/bin/python scripts/cache_aware_proxy.py \
|
|
--combined http://127.0.0.1:8000 http://127.0.0.1:8001 http://127.0.0.1:8002 \
|
|
http://127.0.0.1:8003 http://127.0.0.1:8004 http://127.0.0.1:8005 http://127.0.0.1:8006 \
|
|
--ps-instances http://127.0.0.1:8007 \
|
|
--ps-bootstrap-ports 9005 \
|
|
--bootstrap-ports 8998,8999,9000,9001,9002,9003,9004 \
|
|
--port 9090 > "$OUTDIR/proxy.log" 2>&1 &
|
|
sleep 3
|
|
|
|
echo "Running benchmark..."
|
|
.venv/bin/python -m replayer --trace traces/sampled_1000req_seed42.jsonl \
|
|
--output "$OUTDIR/metrics.jsonl" --endpoint http://localhost:9090 --model "$MODEL" \
|
|
--time-scale 20 --max-inflight-sessions 7 --request-limit 200 -v
|
|
|
|
curl -sf http://localhost:9090/breakdown > "$OUTDIR/breakdown.json" 2>/dev/null || true
|
|
curl -sf http://localhost:9090/stats > "$OUTDIR/stats.json" 2>/dev/null || true
|
|
|
|
# Quick analysis
|
|
.venv/bin/python -c "
|
|
import json
|
|
from collections import Counter
|
|
rows = [json.loads(l) for l in open('$OUTDIR/metrics.jsonl')]
|
|
ok = [r for r in rows if not r.get('error')]
|
|
p = lambda v,q: sorted(v)[min(int(q*len(v)),len(v)-1)] if v else 0
|
|
ttfts = sorted([r['ttft_s'] for r in ok if r.get('ttft_s')])
|
|
tpots = sorted([r['tpot_s'] for r in ok if r.get('tpot_s') and r['tpot_s']>0])
|
|
e2es = sorted([r['latency_s'] for r in ok])
|
|
print('OK=%d/%d TTFT50=%.3f TTFT90=%.3f TPOT90=%.4f E2E50=%.3f' % (
|
|
len(ok), len(rows), p(ttfts,.5), p(ttfts,.9), p(tpots,.9), p(e2es,.5)))
|
|
for lo,hi,cl in [(0,5000,'WARM'),(5000,20000,'MED'),(20000,200000,'HEAVY')]:
|
|
sub = [r for r in ok if lo <= r['input_length'] < hi and r.get('ttft_s')]
|
|
if sub:
|
|
t = sorted([r['ttft_s'] for r in sub])
|
|
tp = sorted([r['tpot_s'] for r in sub if r.get('tpot_s') and r['tpot_s']>0])
|
|
print(' %-6s n=%3d TTFT50=%.3f TPOT90=%.4f' % (cl, len(sub), p(t,.5), p(tp,.9) if tp else 0))
|
|
try:
|
|
bd = json.load(open('$OUTDIR/breakdown.json'))
|
|
rc = Counter(b.get('route_class','') for b in bd)
|
|
print('Routes: %s' % dict(rc))
|
|
ps = [b for b in bd if b.get('route_class') == 'HEAVY_PS']
|
|
if ps:
|
|
pf = [b['t_prefill_done']-b['t_prefill_sent'] for b in ps if b.get('t_prefill_done') and b.get('t_prefill_sent')]
|
|
if pf: print('PS prefill: n=%d p50=%.1fs p90=%.1fs' % (len(pf), p(pf,.5), p(pf,.9)))
|
|
except: pass
|
|
print()
|
|
print('Compare: Phase0A: TTFT50=1.073 TPOT90=0.0738 E2E50=5.096')
|
|
print('Compare: Baseline 8C: TTFT50=1.075 TPOT90=0.0761 E2E50=5.075')
|
|
"
|
|
|
|
cleanup
|
|
echo "=== DONE $(date) ==="
|