feat(agentic): D→P snapshot orchestration in reseed path + CLI flag

Phase 3 — wires the SGLang-side snapshot RPCs (committed in 86412bb)
into the agentic reseed slow-path. On _invoke_kvcache_seeded_router:

  1. POST {prefill_url}/_snapshot/prepare_receive   alloc P-side slots
  2. POST {old_decode_url}/_snapshot/dump           RDMA push session KV
  3. POST {prefill_url}/_snapshot/finalize_ingest   insert into P radix

After step 3 P's radix tree has the session prefix cached; the subsequent
SGLang router-driven prefill on P hits cache instead of re-computing.

Any RPC failure short-circuits to the existing seeded_router fallback
(re-prefill from scratch). All steps are best-effort and structurally
logged for post-hoc analysis.

Flag plumbing:
  cli.py             --enable-d-to-p-sync          (replay + benchmark)
  topology.py        SingleNodeTopology.enable_d_to_p_sync
  stack.py           SGLANG_SNAPSHOT_LINK_ENABLE=1 injection per worker
  replay.py          ReplayConfig.enable_d_to_p_sync +
                     _attempt_d_to_p_sync helper

Snapshot port per worker derives from disaggregation_bootstrap_port +
1000 (set in third_party/.../snapshot/controller.py), so different
workers get distinct mooncake snapshot engines on the same node.

Smoke (next): scripts/smoke_snapshot_sglang_integration.py spawns one
D + one P, exercises the 3 RPCs end-to-end, checks cache_tokens on a
follow-up generate request.

See docs/D_TO_P_SYNC_DESIGN_ZH.md for the full design.
This commit is contained in:
Claude Code Agent
2026-05-13 08:16:46 +08:00
parent 86412bb174
commit b9b0cf0fac
5 changed files with 450 additions and 0 deletions

View File

@@ -0,0 +1,274 @@
#!/usr/bin/env python3
"""End-to-end smoke for the SGLang snapshot link integration.
Brings up TWO SGLang workers on this node (one acts as D, the other as P)
with ``SGLANG_SNAPSHOT_LINK_ENABLE=1`` and exercises the three RPCs:
1. POST {P}/_snapshot/prepare_receive → P allocates kv_pool slots
2. POST {D}/_snapshot/dump → D RDMA-pushes session KV
3. POST {P}/_snapshot/finalize_ingest → P inserts into radix tree
To populate D's SessionAwareCache with a session, we first send a normal
streaming-session generate request to D.
After finalize, we send another generate request to P with the same prefix
and check whether the report says cached_tokens > 0 (cache hit).
This is a minimum-fidelity end-to-end smoke. It does NOT use the full
agentic-pd-hybrid reseed orchestration; that's the next commit.
Required env:
MODEL default /mnt/models/Qwen/Qwen3-30B-A3B-Instruct-2507
Usage:
bash scripts/setup_env.sh && uv run --no-sync python \
scripts/smoke_snapshot_sglang_integration.py
"""
from __future__ import annotations
import argparse
import json
import os
import signal
import subprocess
import sys
import time
from pathlib import Path
from typing import Optional
import httpx
def _build_server_cmd(args, role: str, gpu_id: int, base_port: int,
snapshot_port: int, ib_device: str) -> list:
"""Build the SGLang launch command for one worker (D or P)."""
common = [
sys.executable, "-m", "sglang.launch_server",
"--model-path", args.model,
"--host", "127.0.0.1",
"--port", str(base_port),
"--tp-size", "1",
"--mem-fraction-static", "0.6",
"--disable-cuda-graph",
"--disable-overlap-schedule",
"--enable-streaming-session",
"--disaggregation-mode", role,
"--disaggregation-transfer-backend", "mooncake",
"--disaggregation-bootstrap-port", str(base_port + 5000),
"--disaggregation-ib-device", ib_device,
]
return common
def _server_env(args, gpu_id: int, snapshot_port: int, ib_device: str) -> dict:
env = os.environ.copy()
env["CUDA_VISIBLE_DEVICES"] = str(gpu_id)
env["SGLANG_SNAPSHOT_LINK_ENABLE"] = "1"
env["SGLANG_SNAPSHOT_LINK_HOST"] = "127.0.0.1"
env["SGLANG_SNAPSHOT_LINK_PORT"] = str(snapshot_port)
env["SGLANG_SNAPSHOT_LINK_IB_DEVICE"] = ib_device
env["MOONCAKE_PROTOCOL"] = "rdma"
env["MOONCAKE_DEVICE"] = ib_device
env["MC_TRANSFER_TIMEOUT"] = "1800"
return env
def _wait_for_ready(url: str, timeout: float = 240.0) -> bool:
deadline = time.time() + timeout
while time.time() < deadline:
try:
r = httpx.get(f"{url}/health", timeout=2.0)
if r.status_code == 200:
return True
except Exception:
pass
time.sleep(2)
return False
def main():
ap = argparse.ArgumentParser()
ap.add_argument("--model",
default=os.environ.get("MODEL", "/mnt/models/Qwen/Qwen3-30B-A3B-Instruct-2507"))
ap.add_argument("--d-gpu", type=int, default=1)
ap.add_argument("--p-gpu", type=int, default=0)
ap.add_argument("--d-port", type=int, default=29040)
ap.add_argument("--p-port", type=int, default=29041)
ap.add_argument("--d-snap-port", type=int, default=29045)
ap.add_argument("--p-snap-port", type=int, default=29046)
ap.add_argument("--ib", default="mlx5_60")
ap.add_argument("--log-dir", default="outputs/snapshot_sglang_smoke")
args = ap.parse_args()
log_dir = Path(args.log_dir)
log_dir.mkdir(parents=True, exist_ok=True)
# Spawn P first (so D can find its snapshot endpoint later via prepare_receive)
p_cmd = _build_server_cmd(args, "prefill", args.p_gpu, args.p_port,
args.p_snap_port, args.ib)
p_env = _server_env(args, args.p_gpu, args.p_snap_port, args.ib)
p_stdout = open(log_dir / "p.stdout", "w")
p_stderr = open(log_dir / "p.stderr", "w")
print(f"[smoke] launching P: {' '.join(p_cmd)}")
p_proc = subprocess.Popen(p_cmd, env=p_env, stdout=p_stdout, stderr=p_stderr)
d_cmd = _build_server_cmd(args, "decode", args.d_gpu, args.d_port,
args.d_snap_port, args.ib)
d_env = _server_env(args, args.d_gpu, args.d_snap_port, args.ib)
d_stdout = open(log_dir / "d.stdout", "w")
d_stderr = open(log_dir / "d.stderr", "w")
print(f"[smoke] launching D: {' '.join(d_cmd)}")
d_proc = subprocess.Popen(d_cmd, env=d_env, stdout=d_stdout, stderr=d_stderr)
try:
print(f"[smoke] waiting for P @ 127.0.0.1:{args.p_port} ...")
if not _wait_for_ready(f"http://127.0.0.1:{args.p_port}", timeout=300):
_tail_stderr(log_dir / "p.stderr")
raise RuntimeError("P server did not become healthy")
print(f"[smoke] waiting for D @ 127.0.0.1:{args.d_port} ...")
if not _wait_for_ready(f"http://127.0.0.1:{args.d_port}", timeout=300):
_tail_stderr(log_dir / "d.stderr")
raise RuntimeError("D server did not become healthy")
print(f"[smoke] both servers up — running RPC sanity ...")
session_id = "smoke-sess-001"
# 1. Open streaming session on D
r = httpx.post(f"http://127.0.0.1:{args.d_port}/open_session",
json={"session_id": session_id, "capacity_of_str_len": 8192,
"streaming": True}, timeout=30)
print(f"[smoke] open_session on D → {r.status_code}: {r.text[:200]}")
# 2. Send a small prefill+decode request directly to D (in direct-append mode
# we'd normally go via the pd-router, but for this smoke we send raw)
prompt_ids = [1] * 512 # 512 fake tokens
gen_req = {
"input_ids": prompt_ids,
"sampling_params": {"temperature": 0, "max_new_tokens": 1, "min_new_tokens": 1,
"ignore_eos": True, "skip_special_tokens": False},
"session_params": {"id": session_id},
"stream": False,
}
try:
r = httpx.post(f"http://127.0.0.1:{args.d_port}/generate",
json=gen_req, timeout=60)
print(f"[smoke] D /generate (seed) → {r.status_code}")
except Exception as e:
print(f"[smoke] D /generate failed: {e}")
# 3. Probe snapshot link: prepare_receive on P
num_tokens = 512
prep = httpx.post(
f"http://127.0.0.1:{args.p_port}/_snapshot/prepare_receive",
json={
"session_id": session_id,
"num_tokens": num_tokens,
"expected_bytes_per_layer_k": 0,
"expected_bytes_per_layer_v": 0,
},
timeout=30,
)
print(f"[smoke] prepare_receive on P → {prep.status_code}: {prep.text[:500]}")
if prep.status_code != 200:
return 1
prep_data = prep.json()
if not prep_data.get("ok"):
print(f"[smoke] prepare_receive returned ok=false: {prep_data}")
return 1
# 4. Dump on D
dump = httpx.post(
f"http://127.0.0.1:{args.d_port}/_snapshot/dump",
json={
"session_id": session_id,
"target_snapshot_session_id": prep_data["snapshot_session_id"],
"target_k_base_ptrs": prep_data["k_base_ptrs"],
"target_v_base_ptrs": prep_data["v_base_ptrs"],
"target_slot_indices": prep_data["slot_indices"],
"target_stride_k_bytes": prep_data["stride_k_bytes"],
"target_stride_v_bytes": prep_data["stride_v_bytes"],
"ib_device": args.ib,
},
timeout=60,
)
print(f"[smoke] dump on D → {dump.status_code}: {dump.text[:500]}")
if dump.status_code != 200:
return 1
dump_data = dump.json()
if not dump_data.get("ok"):
print(f"[smoke] dump returned ok=false: {dump_data}")
return 1
print(f"[smoke] dump pushed {dump_data.get('bytes_pushed')} bytes")
# 5. Finalize on P (insert into radix)
fin = httpx.post(
f"http://127.0.0.1:{args.p_port}/_snapshot/finalize_ingest",
json={
"session_id": session_id,
"token_ids": prompt_ids[:num_tokens],
"slot_indices": prep_data["slot_indices"],
},
timeout=30,
)
print(f"[smoke] finalize on P → {fin.status_code}: {fin.text[:500]}")
if fin.status_code != 200:
return 1
fin_data = fin.json()
if not fin_data.get("ok"):
print(f"[smoke] finalize returned ok=false: {fin_data}")
return 1
print(f"[smoke] inserted_prefix_len = {fin_data.get('inserted_prefix_len')}")
# 6. Send the same prefix to P → expect cache hit
gen_p = {
"input_ids": prompt_ids + [42], # prefix + 1 new token
"sampling_params": {"temperature": 0, "max_new_tokens": 1, "min_new_tokens": 1,
"ignore_eos": True, "skip_special_tokens": False},
"stream": False,
}
r = httpx.post(f"http://127.0.0.1:{args.p_port}/generate",
json=gen_p, timeout=60)
print(f"[smoke] P /generate (with cached prefix) → {r.status_code}: "
f"{r.text[:400]}")
try:
body = r.json()
cached = (body.get("meta_info") or {}).get("cached_tokens", 0)
print(f"[smoke] cached_tokens = {cached}")
if cached > 0:
print("[smoke] OVERALL: PASS — P showed cache-hit after snapshot ingest")
return 0
else:
print("[smoke] OVERALL: FAIL — P did not report cache hit")
return 2
except Exception as e:
print(f"[smoke] could not parse P generate response: {e}")
return 3
finally:
for name, proc in [("D", d_proc), ("P", p_proc)]:
try:
proc.send_signal(signal.SIGINT)
except Exception:
pass
for name, proc in [("D", d_proc), ("P", p_proc)]:
try:
proc.wait(timeout=15)
except Exception:
proc.terminate()
try:
proc.wait(timeout=5)
except Exception:
proc.kill()
def _tail_stderr(path: Path, n: int = 60) -> None:
try:
text = path.read_text()
except FileNotFoundError:
return
print(f"--- {path} (last {n}) ---")
for line in text.splitlines()[-n:]:
print(f" {line}")
if __name__ == "__main__":
sys.exit(main())

View File

@@ -283,6 +283,17 @@ def main() -> None:
"See docs/E1_E2_FIX_DESIGN_ZH.md §Q2."
),
)
replay.add_argument(
"--enable-d-to-p-sync",
action="store_true",
help=(
"Enable D→P RDMA KV snapshot push for reseed fast-path. "
"When set, on _invoke_kvcache_seeded_router agentic will probe D's "
"session_aware_cache, RDMA-dump session KV to P's snapshot link, "
"and insert into P's radix tree so the upcoming P prefill hits "
"cache. See docs/D_TO_P_SYNC_DESIGN_ZH.md."
),
)
sample = subparsers.add_parser(
"sample-sessions",
@@ -547,6 +558,14 @@ def main() -> None:
"See docs/E1_E2_FIX_DESIGN_ZH.md §Q2."
),
)
benchmark.add_argument(
"--enable-d-to-p-sync",
action="store_true",
help=(
"Enable D→P RDMA KV snapshot push for reseed fast-path. "
"See docs/D_TO_P_SYNC_DESIGN_ZH.md."
),
)
benchmark.add_argument(
"--sample-profile",
choices=["default", "small-append"],
@@ -634,6 +653,7 @@ def main() -> None:
backpressure_max_pause_s=args.backpressure_max_pause_s,
kvcache_migration_reject_threshold=args.kvcache_migration_reject_threshold,
kvcache_load_floor_bonus=args.kvcache_load_floor_bonus,
enable_d_to_p_sync=args.enable_d_to_p_sync,
)
results = asyncio.run(replay_trace(config))
print(
@@ -876,6 +896,7 @@ def _topology_from_args(args: argparse.Namespace):
force_rdma=args.force_rdma,
trust_remote_code=not args.no_trust_remote_code,
ib_device=args.ib_device,
enable_d_to_p_sync=getattr(args, "enable_d_to_p_sync", False),
prefill_extra_server_args=("--disable-overlap-schedule",),
decode_extra_server_args=("--disable-overlap-schedule",),
direct_extra_server_args=("--enable-streaming-session",),

View File

@@ -116,6 +116,11 @@ class ReplayConfig:
# with shared cross-session prefix. 0 disables. See
# docs/E1_E2_FIX_DESIGN_ZH.md §Q2.
kvcache_load_floor_bonus: int = 0
# D→P snapshot push: when True and reseed fires, agentic will RDMA-dump
# the session's KV from the D-side worker that last held it onto the P
# worker and insert into P's radix tree, so the subsequent P prefill
# hits cache. See docs/D_TO_P_SYNC_DESIGN_ZH.md.
enable_d_to_p_sync: bool = False
structural_log_dir: Path | None = None
@@ -2104,6 +2109,119 @@ async def _invoke_plain_router(
)
async def _attempt_d_to_p_sync(
*,
client: httpx.AsyncClient,
request: TraceRequest,
config: ReplayConfig,
prefill_url: str,
decode_session: DirectSessionState,
) -> dict | None:
"""Try to RDMA-dump session KV from the D that last held it to ``prefill_url``.
Returns a dict with status info on success/skip, or ``None`` on a
non-recoverable error. The caller falls back to normal re-prefill on
any failure.
"""
if not config.enable_d_to_p_sync:
return None
source_d_url = decode_session.server_url
if not source_d_url:
return {"status": "skipped-no-source-d"}
if not decode_session.opened:
return {"status": "skipped-d-closed"}
# Compose token list for radix insert: we don't have the actual token_ids
# on the agentic side in a stable form; use the request's prompt_token_ids
# via the residency bookkeeping. For now we use a length proxy.
target_tokens = max(0, int(_estimate_session_resident_tokens(request)))
if target_tokens <= 0:
return {"status": "skipped-zero-tokens"}
try:
prep_resp = await client.post(
f"{prefill_url}/_snapshot/prepare_receive",
json={
"session_id": request.session_id,
"num_tokens": target_tokens,
},
timeout=30.0,
)
prep_resp.raise_for_status()
prep = prep_resp.json()
except Exception as exc:
return {"status": "prepare-failed", "error": repr(exc)}
if not prep.get("ok"):
return {"status": "prepare-not-ok", "reason": prep.get("reason")}
try:
dump_resp = await client.post(
f"{source_d_url}/_snapshot/dump",
json={
"session_id": request.session_id,
"target_snapshot_session_id": prep["snapshot_session_id"],
"target_k_base_ptrs": prep["k_base_ptrs"],
"target_v_base_ptrs": prep["v_base_ptrs"],
"target_slot_indices": prep["slot_indices"],
"target_stride_k_bytes": prep["stride_k_bytes"],
"target_stride_v_bytes": prep["stride_v_bytes"],
},
timeout=60.0,
)
dump_resp.raise_for_status()
dump = dump_resp.json()
except Exception as exc:
return {"status": "dump-failed", "error": repr(exc)}
if not dump.get("ok"):
return {"status": "dump-not-ok", "reason": dump.get("reason"),
"bytes_pushed": dump.get("bytes_pushed", 0)}
# We need token_ids for radix insert. The caller has request.input_token_ids
# for the first N — use that as best-available approximation.
tokens = list(getattr(request, "input_token_ids", []) or [])
if not tokens:
# No token_ids available — can't insert into radix. P will fall back
# to normal prefill but will have wasted slots. Discard.
try:
await client.post(
f"{prefill_url}/_snapshot/finalize_ingest",
json={
"session_id": request.session_id,
"token_ids": [],
"slot_indices": prep["slot_indices"],
},
timeout=15.0,
)
except Exception:
pass
return {"status": "no-tokens-discard", "bytes_pushed": dump.get("bytes_pushed", 0)}
n = min(len(tokens), len(prep["slot_indices"]))
try:
fin_resp = await client.post(
f"{prefill_url}/_snapshot/finalize_ingest",
json={
"session_id": request.session_id,
"token_ids": tokens[:n],
"slot_indices": prep["slot_indices"][:n],
},
timeout=30.0,
)
fin_resp.raise_for_status()
fin = fin_resp.json()
except Exception as exc:
return {"status": "finalize-failed", "error": repr(exc),
"bytes_pushed": dump.get("bytes_pushed", 0)}
if not fin.get("ok"):
return {"status": "finalize-not-ok", "reason": fin.get("reason"),
"bytes_pushed": dump.get("bytes_pushed", 0)}
return {
"status": "ok",
"bytes_pushed": int(dump.get("bytes_pushed", 0)),
"inserted_prefix_len": int(fin.get("inserted_prefix_len", 0)),
"snapshot_session_id": prep.get("snapshot_session_id"),
}
async def _invoke_kvcache_seeded_router(
*,
client: httpx.AsyncClient,
@@ -2155,6 +2273,31 @@ async def _invoke_kvcache_seeded_router(
decode_session.prefill_server_url = prefill_url
prefill_session_newly_opened = True
# D→P snapshot push (Phase 3) — best-effort; on any failure we silently
# fall back to the existing re-prefill path. The result is logged for
# post-hoc analysis but does not affect correctness.
if config.enable_d_to_p_sync:
sync_result = await _attempt_d_to_p_sync(
client=client,
request=request,
config=config,
prefill_url=prefill_url,
decode_session=decode_session,
)
if sync_result is not None and sync_result.get("status") != "ok":
logger.info(
"d_to_p_sync sid=%s rid=%s skipped: %s",
request.session_id, request.request_id, sync_result,
)
elif sync_result and sync_result.get("status") == "ok":
logger.info(
"d_to_p_sync sid=%s rid=%s pushed=%d ingested_prefix=%d",
request.session_id,
request.request_id,
sync_result.get("bytes_pushed", 0),
sync_result.get("inserted_prefix_len", 0),
)
decode_session_newly_opened = False
try:
prefill_priority = _prefill_priority_for_router_request(

View File

@@ -209,6 +209,15 @@ def _build_process_env(topology: SingleNodeTopology) -> dict[str, str]:
if topology.transfer_backend == "mooncake":
env.setdefault("MC_TRANSFER_TIMEOUT", "1800")
# D→P snapshot link (Phase 2). Each worker reads its own
# `disaggregation_bootstrap_port` and binds at `bootstrap_port + 1000`
# for the snapshot mooncake engine (see
# third_party/sglang/.../disaggregation/snapshot/controller.py).
if topology.enable_d_to_p_sync:
env["SGLANG_SNAPSHOT_LINK_ENABLE"] = "1"
if topology.ib_device:
env.setdefault("SGLANG_SNAPSHOT_LINK_IB_DEVICE", topology.ib_device)
repo_root = Path(__file__).resolve().parents[2]
python_paths = [
str(repo_root / "src"),

View File

@@ -46,6 +46,7 @@ class SingleNodeTopology:
trust_remote_code: bool
force_rdma: bool = False
ib_device: str | None = None
enable_d_to_p_sync: bool = False
extra_server_args: tuple[str, ...] = ()
prefill_extra_server_args: tuple[str, ...] = ()
decode_extra_server_args: tuple[str, ...] = ()
@@ -95,6 +96,7 @@ def build_single_node_topology(
force_rdma: bool = False,
trust_remote_code: bool = True,
ib_device: str | None = None,
enable_d_to_p_sync: bool = False,
extra_server_args: tuple[str, ...] = (),
prefill_extra_server_args: tuple[str, ...] = (),
decode_extra_server_args: tuple[str, ...] = (),
@@ -238,6 +240,7 @@ def build_single_node_topology(
trust_remote_code=trust_remote_code,
force_rdma=force_rdma,
ib_device=ib_device,
enable_d_to_p_sync=enable_d_to_p_sync,
extra_server_args=extra_server_args,
prefill_extra_server_args=prefill_extra_server_args,
decode_extra_server_args=decode_extra_server_args,