📈 feat: add MemoryObserver for GPU-CPU communication tracking
Implement MemoryObserver to track memory transfers between GPU and CPU: - H2D (Host to Device): CPU → GPU transfers - D2H (Device to Host): GPU → CPU transfers - D2D (Device to Device): GPU buffer copies - Supports prefill/decode phase separation Integration points in offload_engine.py: - load_to_slot_layer: H2D with is_prefill parameter - offload_slot_layer_to_cpu, offload_prefill_buffer_async: D2H - write_to_prefill_buffer, write_to_decode_buffer: D2D - load_block_sample_from_cpu, load_block_full_from_cpu: H2D Add bench_offload.py integration for memory stats printing. Benchmark results (Llama-3.1-8B, 64K context): - Full Policy: Prefill H2D 262.13 GB - XAttention: Prefill H2D 386.62 GB (1.48x) Generated with [Claude Code](https://claude.ai/code) via [Happy](https://happy.engineering) Co-Authored-By: Claude <noreply@anthropic.com> Co-Authored-By: Happy <yesreply@happy.engineering>
This commit is contained in:
@@ -31,6 +31,8 @@ Nano-vLLM is a lightweight vLLM implementation (~1,200 lines) for fast offline L
|
|||||||
| [`docs/cpu_offload_optimization_strategies.md`](docs/cpu_offload_optimization_strategies.md) | 🚀 OPT: CPU offload 优化策略:chunk size、CUDA Graph、前沿研究(InfiniGen/ShadowKV) |
|
| [`docs/cpu_offload_optimization_strategies.md`](docs/cpu_offload_optimization_strategies.md) | 🚀 OPT: CPU offload 优化策略:chunk size、CUDA Graph、前沿研究(InfiniGen/ShadowKV) |
|
||||||
| [`docs/gpu_only_xattn_guide.md`](docs/gpu_only_xattn_guide.md) | 🚀 GPU-Only XAttention: 内存预分配、性能分析 (32K +15%, 64K +41%)、CUDA Graph 限制 |
|
| [`docs/gpu_only_xattn_guide.md`](docs/gpu_only_xattn_guide.md) | 🚀 GPU-Only XAttention: 内存预分配、性能分析 (32K +15%, 64K +41%)、CUDA Graph 限制 |
|
||||||
| [`docs/xattn_performance_analysis.md`](docs/xattn_performance_analysis.md) | 📊 XAttention 性能分析: NVTX 标记、block size 影响、estimate vs compute 耗时对比 |
|
| [`docs/xattn_performance_analysis.md`](docs/xattn_performance_analysis.md) | 📊 XAttention 性能分析: NVTX 标记、block size 影响、estimate vs compute 耗时对比 |
|
||||||
|
| [`docs/observer_architecture.md`](docs/observer_architecture.md) | 📊 Observer 架构: InferenceObserver (TTFT/TPOT)、MemoryObserver (H2D/D2H/D2D) 设计 |
|
||||||
|
| [`docs/memory_communication_benchmark.md`](docs/memory_communication_benchmark.md) | 📊 通信量测试: Full vs XAttention 通信量对比 (32K/64K)、阶段分离统计 |
|
||||||
|
|
||||||
## Rules Index
|
## Rules Index
|
||||||
|
|
||||||
|
|||||||
@@ -3,6 +3,14 @@ import time
|
|||||||
from random import randint, seed
|
from random import randint, seed
|
||||||
from nanovllm import LLM, SamplingParams
|
from nanovllm import LLM, SamplingParams
|
||||||
from nanovllm.utils.observer import InferenceObserver
|
from nanovllm.utils.observer import InferenceObserver
|
||||||
|
from nanovllm.utils.memory_observer import MemoryObserver
|
||||||
|
|
||||||
|
|
||||||
|
def print_memory_stats():
|
||||||
|
"""Print MemoryObserver communication statistics"""
|
||||||
|
fmt = MemoryObserver._fmt_bytes
|
||||||
|
print(f"[Memory] Prefill H2D: {fmt(MemoryObserver.prefill_h2d_bytes)}, D2H: {fmt(MemoryObserver.prefill_d2h_bytes)}")
|
||||||
|
print(f" Decode H2D: {fmt(MemoryObserver.decode_h2d_bytes)}, D2H: {fmt(MemoryObserver.decode_d2h_bytes)}")
|
||||||
|
|
||||||
|
|
||||||
def bench_decode(llm, num_seqs, input_len, output_len):
|
def bench_decode(llm, num_seqs, input_len, output_len):
|
||||||
@@ -26,6 +34,7 @@ def bench_decode(llm, num_seqs, input_len, output_len):
|
|||||||
print(f"[Decode] Input: {num_seqs}x{input_len}tok, Output: {decode_tokens}tok, Time: {t:.2f}s")
|
print(f"[Decode] Input: {num_seqs}x{input_len}tok, Output: {decode_tokens}tok, Time: {t:.2f}s")
|
||||||
print(f" TTFT: {ttft_ms:.2f}ms, TPOT: {tpot_ms:.2f}ms")
|
print(f" TTFT: {ttft_ms:.2f}ms, TPOT: {tpot_ms:.2f}ms")
|
||||||
print(f" Decode Throughput: {decode_throughput:.2f} tok/s (from observer)")
|
print(f" Decode Throughput: {decode_throughput:.2f} tok/s (from observer)")
|
||||||
|
print_memory_stats()
|
||||||
|
|
||||||
|
|
||||||
def bench_prefill(llm, num_seqs, input_len):
|
def bench_prefill(llm, num_seqs, input_len):
|
||||||
@@ -51,6 +60,7 @@ def bench_prefill(llm, num_seqs, input_len):
|
|||||||
print(f"[Prefill] Input: {total_input_tokens}tok ({num_seqs}x{input_len})")
|
print(f"[Prefill] Input: {total_input_tokens}tok ({num_seqs}x{input_len})")
|
||||||
print(f" External Time: {t:.2f}s, Throughput: {throughput_external:.2f}tok/s")
|
print(f" External Time: {t:.2f}s, Throughput: {throughput_external:.2f}tok/s")
|
||||||
print(f" Observer TTFT: {ttft_ms:.2f}ms, Throughput: {throughput_observer:.2f}tok/s")
|
print(f" Observer TTFT: {ttft_ms:.2f}ms, Throughput: {throughput_observer:.2f}tok/s")
|
||||||
|
print_memory_stats()
|
||||||
|
|
||||||
|
|
||||||
def main():
|
def main():
|
||||||
@@ -88,6 +98,9 @@ def main():
|
|||||||
path = os.path.expanduser(args.model)
|
path = os.path.expanduser(args.model)
|
||||||
max_len = args.max_len
|
max_len = args.max_len
|
||||||
|
|
||||||
|
# Enable MemoryObserver for communication stats
|
||||||
|
MemoryObserver._enabled = True
|
||||||
|
|
||||||
# Setup policy configuration
|
# Setup policy configuration
|
||||||
if args.enable_quest:
|
if args.enable_quest:
|
||||||
sparse_policy = SparsePolicyType.QUEST
|
sparse_policy = SparsePolicyType.QUEST
|
||||||
|
|||||||
82
docs/memory_communication_benchmark.md
Normal file
82
docs/memory_communication_benchmark.md
Normal file
@@ -0,0 +1,82 @@
|
|||||||
|
# Memory Communication Benchmark
|
||||||
|
|
||||||
|
GPU-CPU 通信量测试结果,对比 Full Policy 和 XAttention BSA Policy。
|
||||||
|
|
||||||
|
## 测试环境
|
||||||
|
|
||||||
|
- **模型**: Llama-3.1-8B-Instruct
|
||||||
|
- **GPU**: RTX 3090 (24GB)
|
||||||
|
- **配置**: `num_gpu_blocks=4`, `block_size=1024`, `enable_cpu_offload=True`
|
||||||
|
- **XAttention 参数**: `threshold=0.95`, `stride=8`
|
||||||
|
|
||||||
|
## 32K 上下文测试结果
|
||||||
|
|
||||||
|
| 指标 | Full Policy | XAttention | 比率 |
|
||||||
|
|------|-------------|------------|------|
|
||||||
|
| **Prefill H2D** | 66.57 GB | 111.12 GB | **1.67x** |
|
||||||
|
| Prefill D2H | 4.29 GB | 4.29 GB | 1.00x |
|
||||||
|
| TTFT | 8473 ms | 10367 ms | 1.22x |
|
||||||
|
|
||||||
|
### XAttention Block Selection (32K)
|
||||||
|
|
||||||
|
| 指标 | 数值 |
|
||||||
|
|------|------|
|
||||||
|
| 可用 blocks | 465 |
|
||||||
|
| 选中 blocks | 374 |
|
||||||
|
| 选择密度 | 80.4% |
|
||||||
|
|
||||||
|
## 64K 上下文测试结果
|
||||||
|
|
||||||
|
| 指标 | Full Policy | XAttention | 比率 |
|
||||||
|
|------|-------------|------------|------|
|
||||||
|
| **Prefill H2D** | 262.13 GB | 386.62 GB | **1.48x** |
|
||||||
|
| Prefill D2H | 8.46 GB | 8.46 GB | 1.00x |
|
||||||
|
| Decode H2D (32 tokens) | 262.13 GB | 262.13 GB | 1.00x |
|
||||||
|
| TTFT | 27081 ms | 33634 ms | 1.24x |
|
||||||
|
|
||||||
|
## 通信量比率对比
|
||||||
|
|
||||||
|
| 上下文长度 | XAttn/Full Prefill H2D 比率 |
|
||||||
|
|------------|----------------------------|
|
||||||
|
| 32K | 1.67x |
|
||||||
|
| 64K | 1.48x |
|
||||||
|
|
||||||
|
### 分析
|
||||||
|
|
||||||
|
1. **XAttention 通信量增加原因**:
|
||||||
|
- Estimate 阶段:加载 **100%** 历史 blocks(用于 attention score 估计)
|
||||||
|
- Compute 阶段:加载 **选中的** blocks(约 70-80%)
|
||||||
|
- 理论比率:`1 + selection_density`
|
||||||
|
|
||||||
|
2. **64K 比率更低的原因**:
|
||||||
|
- 更长上下文时,attention 分布更稀疏
|
||||||
|
- XAttention 的 block 选择更有效(选中比例更低)
|
||||||
|
- First/last block 强制包含的影响相对减小
|
||||||
|
|
||||||
|
3. **Decode 阶段通信量相同**:
|
||||||
|
- XAttention 仅支持 prefill 阶段
|
||||||
|
- Decode 阶段 fallback 到 Full Policy
|
||||||
|
|
||||||
|
## 测试命令
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# 32K Full Policy
|
||||||
|
python bench_offload.py --max-len 32768 --input-len 32000
|
||||||
|
|
||||||
|
# 32K XAttention
|
||||||
|
python bench_offload.py --max-len 32768 --input-len 32000 --enable-xattn
|
||||||
|
|
||||||
|
# 64K Full Policy
|
||||||
|
python bench_offload.py --max-len 65536 --input-len 64000
|
||||||
|
|
||||||
|
# 64K XAttention
|
||||||
|
python bench_offload.py --max-len 65536 --input-len 64000 --enable-xattn
|
||||||
|
|
||||||
|
# 包含 decode 测试
|
||||||
|
python bench_offload.py --max-len 65536 --input-len 64000 --bench-decode --output-len 32
|
||||||
|
```
|
||||||
|
|
||||||
|
## 相关文档
|
||||||
|
|
||||||
|
- [`observer_architecture.md`](observer_architecture.md) - Observer 架构设计
|
||||||
|
- [`xattn_bsa_policy_design.md`](xattn_bsa_policy_design.md) - XAttention BSA 算法设计
|
||||||
194
docs/observer_architecture.md
Normal file
194
docs/observer_architecture.md
Normal file
@@ -0,0 +1,194 @@
|
|||||||
|
# Observer Architecture
|
||||||
|
|
||||||
|
nanovllm 的 Observer 架构用于统计推理过程中的关键指标,采用类变量(class variable)模式实现全局状态管理。
|
||||||
|
|
||||||
|
## 架构概览
|
||||||
|
|
||||||
|
```
|
||||||
|
Observer (基类)
|
||||||
|
├── InferenceObserver - 推理时间指标 (TTFT, TPOT)
|
||||||
|
└── MemoryObserver - 内存传输统计 (H2D, D2H, D2D)
|
||||||
|
```
|
||||||
|
|
||||||
|
## 设计原则
|
||||||
|
|
||||||
|
### 1. 类变量模式
|
||||||
|
|
||||||
|
所有 Observer 使用类变量(而非实例变量)存储状态:
|
||||||
|
|
||||||
|
```python
|
||||||
|
class Observer:
|
||||||
|
"""Observer 基类"""
|
||||||
|
_enabled: bool = True # 类变量,控制是否启用
|
||||||
|
|
||||||
|
class InferenceObserver(Observer):
|
||||||
|
ttft: int = 0 # 类变量,全局共享
|
||||||
|
tpot: int = 0
|
||||||
|
ttft_start: int = 0
|
||||||
|
tpot_start: int = 0
|
||||||
|
```
|
||||||
|
|
||||||
|
**优点**:
|
||||||
|
- 无需实例化,任何地方都可以直接访问
|
||||||
|
- 避免跨模块传递 observer 实例
|
||||||
|
- 适合全局统计场景
|
||||||
|
|
||||||
|
### 2. 启用/禁用控制
|
||||||
|
|
||||||
|
每个 Observer 可独立启用/禁用:
|
||||||
|
|
||||||
|
```python
|
||||||
|
# 启用 MemoryObserver
|
||||||
|
MemoryObserver._enabled = True
|
||||||
|
|
||||||
|
# 禁用后,record_* 方法不会记录
|
||||||
|
MemoryObserver._enabled = False
|
||||||
|
```
|
||||||
|
|
||||||
|
### 3. 阶段分离
|
||||||
|
|
||||||
|
MemoryObserver 支持 prefill/decode 阶段分离统计:
|
||||||
|
|
||||||
|
```python
|
||||||
|
@classmethod
|
||||||
|
def record_h2d(cls, num_bytes: int, is_prefill: bool = True) -> None:
|
||||||
|
if not cls._enabled:
|
||||||
|
return
|
||||||
|
cls.h2d_bytes += num_bytes
|
||||||
|
cls.h2d_count += 1
|
||||||
|
if is_prefill:
|
||||||
|
cls.prefill_h2d_bytes += num_bytes
|
||||||
|
else:
|
||||||
|
cls.decode_h2d_bytes += num_bytes
|
||||||
|
```
|
||||||
|
|
||||||
|
## Observer 实现
|
||||||
|
|
||||||
|
### InferenceObserver
|
||||||
|
|
||||||
|
**位置**: `nanovllm/utils/observer.py`
|
||||||
|
|
||||||
|
**统计指标**:
|
||||||
|
| 指标 | 说明 | 单位 |
|
||||||
|
|------|------|------|
|
||||||
|
| `ttft` | Time To First Token | 纳秒 |
|
||||||
|
| `tpot` | Time Per Output Token | 纳秒 |
|
||||||
|
| `ttft_start` | TTFT 计时开始点 | 纳秒 |
|
||||||
|
| `tpot_start` | TPOT 计时开始点 | 纳秒 |
|
||||||
|
|
||||||
|
**统计位置**:
|
||||||
|
| 位置 | 代码 | 说明 |
|
||||||
|
|------|------|------|
|
||||||
|
| `scheduler.py:add()` | `InferenceObserver.ttft_start = perf_counter_ns()` | 开始计时 |
|
||||||
|
| `llm_engine.py:step()` | `InferenceObserver.ttft = ... - ttft_start` | Prefill 完成后计算 TTFT |
|
||||||
|
| `llm_engine.py:step()` | `InferenceObserver.tpot = ... - tpot_start` | Decode 时计算 TPOT |
|
||||||
|
|
||||||
|
### MemoryObserver
|
||||||
|
|
||||||
|
**位置**: `nanovllm/utils/memory_observer.py`
|
||||||
|
|
||||||
|
**统计指标**:
|
||||||
|
| 指标 | 说明 |
|
||||||
|
|------|------|
|
||||||
|
| `h2d_bytes` / `h2d_count` | Host to Device 传输量/次数 |
|
||||||
|
| `d2h_bytes` / `d2h_count` | Device to Host 传输量/次数 |
|
||||||
|
| `d2d_bytes` / `d2d_count` | Device to Device 复制量/次数 |
|
||||||
|
| `prefill_h2d_bytes` / `prefill_d2h_bytes` | Prefill 阶段 H2D/D2H |
|
||||||
|
| `decode_h2d_bytes` / `decode_d2h_bytes` | Decode 阶段 H2D/D2H |
|
||||||
|
|
||||||
|
**统计位置** (均在 `offload_engine.py`):
|
||||||
|
|
||||||
|
| 方法 | 传输类型 | 说明 |
|
||||||
|
|------|----------|------|
|
||||||
|
| `load_to_slot_layer()` | H2D | 从 CPU 加载 block 到 GPU slot |
|
||||||
|
| `load_block_sample_from_cpu()` | H2D | 采样加载(Quest) |
|
||||||
|
| `load_block_full_from_cpu()` | H2D | 完整加载 block |
|
||||||
|
| `offload_slot_layer_to_cpu()` | D2H | GPU slot 卸载到 CPU |
|
||||||
|
| `offload_prefill_buffer_async()` | D2H | Prefill buffer 异步卸载 |
|
||||||
|
| `write_to_prefill_buffer()` | D2D | 写入 prefill buffer |
|
||||||
|
| `write_to_decode_buffer()` | D2D | 写入 decode buffer |
|
||||||
|
|
||||||
|
**重置位置**:
|
||||||
|
| 位置 | 代码 |
|
||||||
|
|------|------|
|
||||||
|
| `llm_engine.py:generate()` | `MemoryObserver.complete_reset()` |
|
||||||
|
| `llm_engine.py:generate()` | `InferenceObserver.complete_reset()` |
|
||||||
|
|
||||||
|
## 使用示例
|
||||||
|
|
||||||
|
### 1. 启用并统计
|
||||||
|
|
||||||
|
```python
|
||||||
|
from nanovllm.utils.memory_observer import MemoryObserver
|
||||||
|
|
||||||
|
# 启用统计
|
||||||
|
MemoryObserver._enabled = True
|
||||||
|
|
||||||
|
# 运行推理
|
||||||
|
outputs = llm.generate(prompts, sampling_params)
|
||||||
|
|
||||||
|
# 获取结果
|
||||||
|
print(f"Prefill H2D: {MemoryObserver.prefill_h2d_bytes / 1e9:.2f} GB")
|
||||||
|
print(f"Decode H2D: {MemoryObserver.decode_h2d_bytes / 1e9:.2f} GB")
|
||||||
|
|
||||||
|
# 或使用 print_summary
|
||||||
|
MemoryObserver.print_summary()
|
||||||
|
```
|
||||||
|
|
||||||
|
### 2. 在 bench_offload.py 中
|
||||||
|
|
||||||
|
```python
|
||||||
|
from nanovllm.utils.memory_observer import MemoryObserver
|
||||||
|
|
||||||
|
# 启用
|
||||||
|
MemoryObserver._enabled = True
|
||||||
|
|
||||||
|
# benchmark 结束后打印
|
||||||
|
def print_memory_stats():
|
||||||
|
fmt = MemoryObserver._fmt_bytes
|
||||||
|
print(f"[Memory] Prefill H2D: {fmt(MemoryObserver.prefill_h2d_bytes)}")
|
||||||
|
print(f" Decode H2D: {fmt(MemoryObserver.decode_h2d_bytes)}")
|
||||||
|
```
|
||||||
|
|
||||||
|
### 3. 获取结构化数据
|
||||||
|
|
||||||
|
```python
|
||||||
|
summary = MemoryObserver.get_summary()
|
||||||
|
# {
|
||||||
|
# "total": {"h2d_bytes": ..., "d2h_bytes": ..., "d2d_bytes": ...},
|
||||||
|
# "prefill": {"h2d_bytes": ..., "d2h_bytes": ...},
|
||||||
|
# "decode": {"h2d_bytes": ..., "d2h_bytes": ...}
|
||||||
|
# }
|
||||||
|
```
|
||||||
|
|
||||||
|
## 添加新 Observer
|
||||||
|
|
||||||
|
1. 继承 `Observer` 基类
|
||||||
|
2. 定义类变量存储统计数据
|
||||||
|
3. 实现 `record_*` 方法(需检查 `_enabled`)
|
||||||
|
4. 实现 `complete_reset()` 方法
|
||||||
|
5. 在相关代码位置添加 `record_*` 调用
|
||||||
|
6. 在 `llm_engine.py:generate()` 中添加 reset 调用
|
||||||
|
|
||||||
|
```python
|
||||||
|
from nanovllm.utils.observer import Observer
|
||||||
|
|
||||||
|
class MyObserver(Observer):
|
||||||
|
_enabled: bool = False
|
||||||
|
my_metric: int = 0
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def record_event(cls, value: int) -> None:
|
||||||
|
if not cls._enabled:
|
||||||
|
return
|
||||||
|
cls.my_metric += value
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def complete_reset(cls) -> None:
|
||||||
|
cls.my_metric = 0
|
||||||
|
```
|
||||||
|
|
||||||
|
## 相关文档
|
||||||
|
|
||||||
|
- [`memory_communication_benchmark.md`](memory_communication_benchmark.md) - 通信量测试结果
|
||||||
|
- [`architecture_guide.md`](architecture_guide.md) - 整体架构指南
|
||||||
@@ -11,6 +11,7 @@ from nanovllm.engine.sequence import Sequence
|
|||||||
from nanovllm.engine.scheduler import Scheduler
|
from nanovllm.engine.scheduler import Scheduler
|
||||||
from nanovllm.engine.model_runner import ModelRunner
|
from nanovllm.engine.model_runner import ModelRunner
|
||||||
from nanovllm.utils.observer import InferenceObserver
|
from nanovllm.utils.observer import InferenceObserver
|
||||||
|
from nanovllm.utils.memory_observer import MemoryObserver
|
||||||
|
|
||||||
|
|
||||||
class LLMEngine:
|
class LLMEngine:
|
||||||
@@ -95,6 +96,7 @@ class LLMEngine:
|
|||||||
debug_enabled = log_level.upper() == 'DEBUG'
|
debug_enabled = log_level.upper() == 'DEBUG'
|
||||||
|
|
||||||
InferenceObserver.complete_reset()
|
InferenceObserver.complete_reset()
|
||||||
|
MemoryObserver.complete_reset()
|
||||||
if use_tqdm:
|
if use_tqdm:
|
||||||
pbar = tqdm(total=len(prompts), desc="Generating", dynamic_ncols=True)
|
pbar = tqdm(total=len(prompts), desc="Generating", dynamic_ncols=True)
|
||||||
if not isinstance(sampling_params, list):
|
if not isinstance(sampling_params, list):
|
||||||
|
|||||||
@@ -17,6 +17,7 @@ from dataclasses import dataclass
|
|||||||
from nanovllm.kvcache.kernels import gathered_copy_kv
|
from nanovllm.kvcache.kernels import gathered_copy_kv
|
||||||
from nanovllm.comm import memcpy_2d_async
|
from nanovllm.comm import memcpy_2d_async
|
||||||
from nanovllm.utils.logger import get_logger
|
from nanovllm.utils.logger import get_logger
|
||||||
|
from nanovllm.utils.memory_observer import MemoryObserver
|
||||||
|
|
||||||
# Import for type hints only (avoid circular import)
|
# Import for type hints only (avoid circular import)
|
||||||
from typing import TYPE_CHECKING
|
from typing import TYPE_CHECKING
|
||||||
@@ -376,7 +377,8 @@ class OffloadEngine:
|
|||||||
self.ring_slot_compute_done[slot_idx].record()
|
self.ring_slot_compute_done[slot_idx].record()
|
||||||
|
|
||||||
def load_to_slot_layer(
|
def load_to_slot_layer(
|
||||||
self, slot_idx: int, layer_id: int, cpu_block_id: int, chunk_idx: int = -1
|
self, slot_idx: int, layer_id: int, cpu_block_id: int, chunk_idx: int = -1,
|
||||||
|
is_prefill: bool = True,
|
||||||
) -> None:
|
) -> None:
|
||||||
"""
|
"""
|
||||||
Async load a single CPU block to a ring buffer slot for one layer.
|
Async load a single CPU block to a ring buffer slot for one layer.
|
||||||
@@ -393,6 +395,7 @@ class OffloadEngine:
|
|||||||
layer_id: Layer index to load (for CPU cache indexing)
|
layer_id: Layer index to load (for CPU cache indexing)
|
||||||
cpu_block_id: Source CPU block ID
|
cpu_block_id: Source CPU block ID
|
||||||
chunk_idx: Optional chunk index for NVTX labeling (-1 means not specified)
|
chunk_idx: Optional chunk index for NVTX labeling (-1 means not specified)
|
||||||
|
is_prefill: True if in prefill phase, False if in decode phase (for MemoryObserver)
|
||||||
"""
|
"""
|
||||||
logger.debug(f"Ring load: layer={layer_id}, CPU[{cpu_block_id}] -> GPU slot[{slot_idx}]")
|
logger.debug(f"Ring load: layer={layer_id}, CPU[{cpu_block_id}] -> GPU slot[{slot_idx}]")
|
||||||
|
|
||||||
@@ -425,6 +428,9 @@ class OffloadEngine:
|
|||||||
self.ring_slot_ready[slot_idx].record(stream)
|
self.ring_slot_ready[slot_idx].record(stream)
|
||||||
nvtx.pop_range()
|
nvtx.pop_range()
|
||||||
|
|
||||||
|
# Record H2D transfer: K + V = 2 * block_bytes
|
||||||
|
MemoryObserver.record_h2d(2 * self.gpu_block_bytes, is_prefill=is_prefill)
|
||||||
|
|
||||||
def wait_slot_layer(self, slot_idx: int) -> None:
|
def wait_slot_layer(self, slot_idx: int) -> None:
|
||||||
"""
|
"""
|
||||||
Wait for a slot's loading to complete.
|
Wait for a slot's loading to complete.
|
||||||
@@ -499,6 +505,9 @@ class OffloadEngine:
|
|||||||
self.ring_slot_offload_done[slot_idx].record(self.transfer_stream_main)
|
self.ring_slot_offload_done[slot_idx].record(self.transfer_stream_main)
|
||||||
nvtx.pop_range()
|
nvtx.pop_range()
|
||||||
|
|
||||||
|
# Record D2H transfer: K + V = 2 * block_bytes
|
||||||
|
MemoryObserver.record_d2h(2 * self.gpu_block_bytes, is_prefill=is_prefill)
|
||||||
|
|
||||||
# ----- KV access methods for ring buffer -----
|
# ----- KV access methods for ring buffer -----
|
||||||
|
|
||||||
def get_kv_for_slot(self, slot_idx: int) -> Tuple[Tensor, Tensor]:
|
def get_kv_for_slot(self, slot_idx: int) -> Tuple[Tensor, Tensor]:
|
||||||
@@ -745,6 +754,10 @@ class OffloadEngine:
|
|||||||
self.prefill_v_buffer[layer_id, :num_tokens].copy_(v)
|
self.prefill_v_buffer[layer_id, :num_tokens].copy_(v)
|
||||||
torch.cuda.nvtx.range_pop()
|
torch.cuda.nvtx.range_pop()
|
||||||
|
|
||||||
|
# Record D2D transfer: K + V
|
||||||
|
transfer_bytes = 2 * k.numel() * k.element_size()
|
||||||
|
MemoryObserver.record_d2d(transfer_bytes)
|
||||||
|
|
||||||
def write_to_decode_buffer(
|
def write_to_decode_buffer(
|
||||||
self,
|
self,
|
||||||
layer_id: int,
|
layer_id: int,
|
||||||
@@ -768,6 +781,10 @@ class OffloadEngine:
|
|||||||
self.decode_v_buffer[layer_id, pos_in_block].copy_(v)
|
self.decode_v_buffer[layer_id, pos_in_block].copy_(v)
|
||||||
torch.cuda.nvtx.range_pop()
|
torch.cuda.nvtx.range_pop()
|
||||||
|
|
||||||
|
# Record D2D transfer: K + V (single token)
|
||||||
|
transfer_bytes = 2 * k.numel() * k.element_size()
|
||||||
|
MemoryObserver.record_d2d(transfer_bytes)
|
||||||
|
|
||||||
def offload_prefill_buffer_async(
|
def offload_prefill_buffer_async(
|
||||||
self,
|
self,
|
||||||
layer_id: int,
|
layer_id: int,
|
||||||
@@ -813,6 +830,9 @@ class OffloadEngine:
|
|||||||
self.prefill_offload_events[layer_id].record(stream)
|
self.prefill_offload_events[layer_id].record(stream)
|
||||||
nvtx.pop_range()
|
nvtx.pop_range()
|
||||||
|
|
||||||
|
# Record D2H transfer: K + V = 2 * block_bytes
|
||||||
|
MemoryObserver.record_d2h(2 * self.gpu_block_bytes, is_prefill=True)
|
||||||
|
|
||||||
def wait_all_prefill_offloads(self) -> None:
|
def wait_all_prefill_offloads(self) -> None:
|
||||||
"""Wait for all prefill buffer offloads to complete."""
|
"""Wait for all prefill buffer offloads to complete."""
|
||||||
for stream in self.prefill_offload_streams:
|
for stream in self.prefill_offload_streams:
|
||||||
@@ -851,6 +871,11 @@ class OffloadEngine:
|
|||||||
v_sample = self.v_cache_cpu[
|
v_sample = self.v_cache_cpu[
|
||||||
layer_id, cpu_block_id, :num_samples
|
layer_id, cpu_block_id, :num_samples
|
||||||
].clone().cuda()
|
].clone().cuda()
|
||||||
|
|
||||||
|
# Record H2D transfer: K + V samples
|
||||||
|
transfer_bytes = 2 * k_sample.numel() * k_sample.element_size()
|
||||||
|
MemoryObserver.record_h2d(transfer_bytes, is_prefill=True)
|
||||||
|
|
||||||
return k_sample, v_sample
|
return k_sample, v_sample
|
||||||
|
|
||||||
def load_block_full_from_cpu(
|
def load_block_full_from_cpu(
|
||||||
@@ -877,4 +902,8 @@ class OffloadEngine:
|
|||||||
v_full = self.v_cache_cpu[
|
v_full = self.v_cache_cpu[
|
||||||
layer_id, cpu_block_id
|
layer_id, cpu_block_id
|
||||||
].clone().cuda()
|
].clone().cuda()
|
||||||
|
|
||||||
|
# Record H2D transfer: K + V full block
|
||||||
|
MemoryObserver.record_h2d(2 * self.gpu_block_bytes, is_prefill=True)
|
||||||
|
|
||||||
return k_full, v_full
|
return k_full, v_full
|
||||||
|
|||||||
@@ -422,7 +422,7 @@ class FullAttentionPolicy(SparsePolicy):
|
|||||||
num_preload = min(num_slots, num_blocks)
|
num_preload = min(num_slots, num_blocks)
|
||||||
for i in range(num_preload):
|
for i in range(num_preload):
|
||||||
cpu_block_id = cpu_block_table[i]
|
cpu_block_id = cpu_block_table[i]
|
||||||
offload_engine.load_to_slot_layer(load_slots[i], layer_id, cpu_block_id, chunk_idx=cpu_block_id)
|
offload_engine.load_to_slot_layer(load_slots[i], layer_id, cpu_block_id, chunk_idx=cpu_block_id, is_prefill=False)
|
||||||
|
|
||||||
# Phase 2: Process blocks with pipeline
|
# Phase 2: Process blocks with pipeline
|
||||||
for block_idx in range(num_blocks):
|
for block_idx in range(num_blocks):
|
||||||
@@ -456,7 +456,7 @@ class FullAttentionPolicy(SparsePolicy):
|
|||||||
next_block_idx = block_idx + num_slots
|
next_block_idx = block_idx + num_slots
|
||||||
if next_block_idx < num_blocks:
|
if next_block_idx < num_blocks:
|
||||||
next_cpu_block_id = cpu_block_table[next_block_idx]
|
next_cpu_block_id = cpu_block_table[next_block_idx]
|
||||||
offload_engine.load_to_slot_layer(current_slot, layer_id, next_cpu_block_id, chunk_idx=next_cpu_block_id)
|
offload_engine.load_to_slot_layer(current_slot, layer_id, next_cpu_block_id, chunk_idx=next_cpu_block_id, is_prefill=False)
|
||||||
|
|
||||||
# Merge with accumulated
|
# Merge with accumulated
|
||||||
with torch.cuda.stream(compute_stream):
|
with torch.cuda.stream(compute_stream):
|
||||||
|
|||||||
133
nanovllm/utils/memory_observer.py
Normal file
133
nanovllm/utils/memory_observer.py
Normal file
@@ -0,0 +1,133 @@
|
|||||||
|
"""
|
||||||
|
MemoryObserver - 内存传输统计 Observer。
|
||||||
|
|
||||||
|
统计 GPU-CPU 间的数据传输量:
|
||||||
|
- H2D (Host to Device): CPU → GPU
|
||||||
|
- D2H (Device to Host): GPU → CPU
|
||||||
|
- D2D (Device to Device): GPU → GPU (buffer copy)
|
||||||
|
"""
|
||||||
|
|
||||||
|
from nanovllm.utils.observer import Observer
|
||||||
|
|
||||||
|
|
||||||
|
class MemoryObserver(Observer):
|
||||||
|
"""
|
||||||
|
内存传输 Observer,统计 GPU-CPU 间的数据传输量。
|
||||||
|
|
||||||
|
统计类型:
|
||||||
|
- H2D (Host to Device): CPU → GPU
|
||||||
|
- D2H (Device to Host): GPU → CPU
|
||||||
|
- D2D (Device to Device): GPU → GPU (buffer copy)
|
||||||
|
|
||||||
|
统计位置(均在 offload_engine.py):
|
||||||
|
- H2D: load_to_slot_layer(), load_block_sample_from_cpu(), load_block_full_from_cpu()
|
||||||
|
- D2H: offload_slot_layer_to_cpu(), offload_prefill_buffer_async()
|
||||||
|
- D2D: write_to_prefill_buffer(), write_to_decode_buffer()
|
||||||
|
- 重置: llm_engine.py:generate() - 与 InferenceObserver 一起重置
|
||||||
|
"""
|
||||||
|
|
||||||
|
_enabled: bool = False # 默认禁用,需要显式启用
|
||||||
|
|
||||||
|
# H2D 统计
|
||||||
|
h2d_bytes: int = 0
|
||||||
|
h2d_count: int = 0
|
||||||
|
|
||||||
|
# D2H 统计
|
||||||
|
d2h_bytes: int = 0
|
||||||
|
d2h_count: int = 0
|
||||||
|
|
||||||
|
# D2D 统计
|
||||||
|
d2d_bytes: int = 0
|
||||||
|
d2d_count: int = 0
|
||||||
|
|
||||||
|
# 按阶段统计
|
||||||
|
prefill_h2d_bytes: int = 0
|
||||||
|
prefill_d2h_bytes: int = 0
|
||||||
|
decode_h2d_bytes: int = 0
|
||||||
|
decode_d2h_bytes: int = 0
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def record_h2d(cls, num_bytes: int, is_prefill: bool = True) -> None:
|
||||||
|
"""记录 H2D 传输"""
|
||||||
|
if not cls._enabled:
|
||||||
|
return
|
||||||
|
cls.h2d_bytes += num_bytes
|
||||||
|
cls.h2d_count += 1
|
||||||
|
if is_prefill:
|
||||||
|
cls.prefill_h2d_bytes += num_bytes
|
||||||
|
else:
|
||||||
|
cls.decode_h2d_bytes += num_bytes
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def record_d2h(cls, num_bytes: int, is_prefill: bool = True) -> None:
|
||||||
|
"""记录 D2H 传输"""
|
||||||
|
if not cls._enabled:
|
||||||
|
return
|
||||||
|
cls.d2h_bytes += num_bytes
|
||||||
|
cls.d2h_count += 1
|
||||||
|
if is_prefill:
|
||||||
|
cls.prefill_d2h_bytes += num_bytes
|
||||||
|
else:
|
||||||
|
cls.decode_d2h_bytes += num_bytes
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def record_d2d(cls, num_bytes: int) -> None:
|
||||||
|
"""记录 D2D 传输"""
|
||||||
|
if not cls._enabled:
|
||||||
|
return
|
||||||
|
cls.d2d_bytes += num_bytes
|
||||||
|
cls.d2d_count += 1
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def complete_reset(cls) -> None:
|
||||||
|
"""重置所有统计"""
|
||||||
|
cls.h2d_bytes = cls.h2d_count = 0
|
||||||
|
cls.d2h_bytes = cls.d2h_count = 0
|
||||||
|
cls.d2d_bytes = cls.d2d_count = 0
|
||||||
|
cls.prefill_h2d_bytes = cls.prefill_d2h_bytes = 0
|
||||||
|
cls.decode_h2d_bytes = cls.decode_d2h_bytes = 0
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def get_summary(cls) -> dict:
|
||||||
|
"""返回统计摘要"""
|
||||||
|
return {
|
||||||
|
"total": {
|
||||||
|
"h2d_bytes": cls.h2d_bytes,
|
||||||
|
"h2d_count": cls.h2d_count,
|
||||||
|
"d2h_bytes": cls.d2h_bytes,
|
||||||
|
"d2h_count": cls.d2h_count,
|
||||||
|
"d2d_bytes": cls.d2d_bytes,
|
||||||
|
"d2d_count": cls.d2d_count,
|
||||||
|
},
|
||||||
|
"prefill": {
|
||||||
|
"h2d_bytes": cls.prefill_h2d_bytes,
|
||||||
|
"d2h_bytes": cls.prefill_d2h_bytes,
|
||||||
|
},
|
||||||
|
"decode": {
|
||||||
|
"h2d_bytes": cls.decode_h2d_bytes,
|
||||||
|
"d2h_bytes": cls.decode_d2h_bytes,
|
||||||
|
},
|
||||||
|
}
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def _fmt_bytes(cls, b: int) -> str:
|
||||||
|
"""格式化字节数"""
|
||||||
|
if b >= 1e9:
|
||||||
|
return f"{b/1e9:.2f} GB"
|
||||||
|
if b >= 1e6:
|
||||||
|
return f"{b/1e6:.2f} MB"
|
||||||
|
if b >= 1e3:
|
||||||
|
return f"{b/1e3:.2f} KB"
|
||||||
|
return f"{b} B"
|
||||||
|
|
||||||
|
@classmethod
|
||||||
|
def print_summary(cls) -> None:
|
||||||
|
"""打印人类可读的摘要"""
|
||||||
|
fmt = cls._fmt_bytes
|
||||||
|
total = cls.h2d_bytes + cls.d2h_bytes + cls.d2d_bytes
|
||||||
|
print(f"[MemoryObserver] Total: {fmt(total)}")
|
||||||
|
print(f" H2D: {fmt(cls.h2d_bytes)} ({cls.h2d_count} ops)")
|
||||||
|
print(f" D2H: {fmt(cls.d2h_bytes)} ({cls.d2h_count} ops)")
|
||||||
|
print(f" D2D: {fmt(cls.d2d_bytes)} ({cls.d2d_count} ops)")
|
||||||
|
print(f" Prefill - H2D: {fmt(cls.prefill_h2d_bytes)}, D2H: {fmt(cls.prefill_d2h_bytes)}")
|
||||||
|
print(f" Decode - H2D: {fmt(cls.decode_h2d_bytes)}, D2H: {fmt(cls.decode_d2h_bytes)}")
|
||||||
Reference in New Issue
Block a user