[OpenVINO] Enable GPU support for OpenVINO vLLM backend (#8192)

docs/source/getting_started/openvino-installation.rst

@@ -3,7 +3,7 @@
 Installation with OpenVINO
 ==========================
 
-vLLM powered by OpenVINO supports all LLM models from :doc:`vLLM supported models list <../models/supported_models>` and can perform optimal model serving on all x86-64 CPUs with, at least, AVX2 support. OpenVINO vLLM backend supports the following advanced vLLM features:
+vLLM powered by OpenVINO supports all LLM models from :doc:`vLLM supported models list <../models/supported_models>` and can perform optimal model serving on all x86-64 CPUs with, at least, AVX2 support, as well as on both integrated and discrete Intel® GPUs (`the list of supported GPUs <https://docs.openvino.ai/2024/about-openvino/release-notes-openvino/system-requirements.html#gpu>`_). OpenVINO vLLM backend supports the following advanced vLLM features:
 
 - Prefix caching (``--enable-prefix-caching``)
 - Chunked prefill (``--enable-chunked-prefill``)
@@ -53,34 +53,57 @@ Install from source
       $ pip install --upgrade pip
       $ pip install -r requirements-build.txt --extra-index-url https://download.pytorch.org/whl/cpu
 
-- Finally, install vLLM with OpenVINO backend: 
+- Finally, install vLLM with OpenVINO backend:
 
   .. code-block:: console
 
       $ PIP_EXTRA_INDEX_URL="https://download.pytorch.org/whl/cpu" VLLM_TARGET_DEVICE=openvino python -m pip install -v .
 
+- [Optional] To use vLLM OpenVINO backend with a GPU device, ensure your system is properly set up. Follow the instructions provided here: `https://docs.openvino.ai/2024/get-started/configurations/configurations-intel-gpu.html <https://docs.openvino.ai/2024/get-started/configurations/configurations-intel-gpu.html>`_.
+
 .. _openvino_backend_performance_tips:
 
 Performance tips
 ----------------
 
-vLLM OpenVINO backend uses the following environment variables to control behavior:
+vLLM OpenVINO backend environment variables
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+- ``VLLM_OPENVINO_DEVICE`` to specify which device utilize for the inference. If there are multiple GPUs in the system, additional indexes can be used to choose the proper one (e.g, ``VLLM_OPENVINO_DEVICE=GPU.1``). If the value is not specified, CPU device is used by default.
+
+- ``VLLM_OPENVINO_ENABLE_QUANTIZED_WEIGHTS=ON`` to enable U8 weights compression during model loading stage. By default, compression is turned off. You can also export model with different compression techniques using `optimum-cli` and pass exported folder as `<model_id>`
+
+CPU performance tips
+~~~~~~~~~~~~~~~~~~~~
+
+CPU uses the following environment variables to control behavior:
 
 - ``VLLM_OPENVINO_KVCACHE_SPACE`` to specify the KV Cache size (e.g, ``VLLM_OPENVINO_KVCACHE_SPACE=40`` means 40 GB space for KV cache), larger setting will allow vLLM running more requests in parallel. This parameter should be set based on the hardware configuration and memory management pattern of users.
 
 - ``VLLM_OPENVINO_CPU_KV_CACHE_PRECISION=u8`` to control KV cache precision. By default, FP16 / BF16 is used depending on platform.
 
-- ``VLLM_OPENVINO_ENABLE_QUANTIZED_WEIGHTS=ON`` to enable U8 weights compression during model loading stage. By default, compression is turned off. You can also export model with different compression techniques using `optimum-cli` and pass exported folder as `<model_id>`
-
 To enable better TPOT / TTFT latency, you can use vLLM's chunked prefill feature (``--enable-chunked-prefill``). Based on the experiments, the recommended batch size is ``256`` (``--max-num-batched-tokens``)
 
-OpenVINO best known configuration is:
+OpenVINO best known configuration for CPU is:
 
 .. code-block:: console
 
     $ VLLM_OPENVINO_KVCACHE_SPACE=100 VLLM_OPENVINO_CPU_KV_CACHE_PRECISION=u8 VLLM_OPENVINO_ENABLE_QUANTIZED_WEIGHTS=ON \
         python3 vllm/benchmarks/benchmark_throughput.py --model meta-llama/Llama-2-7b-chat-hf --dataset vllm/benchmarks/ShareGPT_V3_unfiltered_cleaned_split.json --enable-chunked-prefill --max-num-batched-tokens 256
 
+GPU performance tips
+~~~~~~~~~~~~~~~~~~~~
+GPU device implements the logic for automatic detection of available GPU memory and, by default, tries to reserve as much memory as possible for the KV cache (taking into account ``gpu_memory_utilization`` option). However, this behavior can be overridden by explicitly specifying the desired amount of memory for the KV cache using ``VLLM_OPENVINO_KVCACHE_SPACE`` environment variable (e.g, ``VLLM_OPENVINO_KVCACHE_SPACE=8`` means 8 GB space for KV cache).
+
+Currently, the best performance using GPU can be achieved with the default vLLM execution parameters for models with quantized weights (8 and 4-bit integer data types are supported) and `preemption-mode=swap`.
+
+OpenVINO best known configuration for GPU is:
+
+.. code-block:: console
+
+    $ VLLM_OPENVINO_DEVICE=GPU VLLM_OPENVINO_ENABLE_QUANTIZED_WEIGHTS=ON \
+        python3 vllm/benchmarks/benchmark_throughput.py --model meta-llama/Llama-2-7b-chat-hf --dataset vllm/benchmarks/ShareGPT_V3_unfiltered_cleaned_split.json
+
 .. _openvino_backend_limitations:
 
 Limitations

requirements-openvino.txt

@@ -3,5 +3,6 @@
 
 # OpenVINO dependencies
 torch >= 2.1.2
-openvino ~= 2024.3.0
-optimum-intel[openvino] >= 1.18.2
+openvino ~= 2024.4.0
+openvino-tokenizers[transformers] ~= 2024.4.0
+optimum-intel[openvino] >= 1.19.0

vllm/attention/backends/openvino.py

@@ -9,6 +9,31 @@
 from vllm.attention.backends.utils import CommonAttentionState
 
 
+def copy_cache_block(src_tensor: ov.Tensor, dst_tensor: ov.Tensor,
+                     src_offset: int, dst_offset: int) -> None:
+
+    def create_roi_tensor(
+        tensor: ov.Tensor,
+        block_number: int,
+    ) -> ov.Tensor:
+        roi_begin = ov.runtime.Coordinate([0, 0, 0, 0])
+        roi_end = ov.runtime.Coordinate(tensor.get_shape())
+
+        roi_begin[0] = block_number
+        roi_end[0] = block_number + 1
+
+        if isinstance(tensor, ov.Tensor):
+            return ov.Tensor(tensor, roi_begin, roi_end)
+        else:
+            return ov.RemoteTensor(tensor, roi_begin, roi_end)
+
+    src_roi_tensor = \
+        create_roi_tensor(src_tensor, src_offset)
+    dst_roi_tensor = \
+        create_roi_tensor(dst_tensor, dst_offset)
+    src_roi_tensor.copy_to(dst_roi_tensor)
+
+
 class OpenVINOAttentionBackend(AttentionBackend):
 
     @staticmethod
@@ -44,13 +69,12 @@ def get_kv_cache_shape(
 
     @staticmethod
     def swap_blocks(
-        src_kv_cache: ov.Tensor,
-        dst_kv_cache: ov.Tensor,
-        src_to_dst: torch.Tensor,
+        src_tensor: ov.Tensor,
+        dst_tensor: ov.Tensor,
+        src_to_dists: List[Tuple[int, int]],
     ) -> None:
-        # OpenVINO currently supports only CPU, which does not require
-        # swap of KV cache blocks
-        raise NotImplementedError
+        for src, dst in src_to_dists:
+            copy_cache_block(src_tensor, dst_tensor, src, dst)
 
     @staticmethod
     def copy_blocks(
@@ -59,8 +83,8 @@ def copy_blocks(
     ) -> None:
         for src, dst in src_to_dists:
             for key_cache, value_cache in kv_caches:
-                key_cache.data[dst, :] = key_cache.data[src, :]
-                value_cache.data[dst, :] = value_cache.data[src, :]
+                copy_cache_block(key_cache, key_cache, src, dst)
+                copy_cache_block(value_cache, value_cache, src, dst)
 
 
 @dataclass

vllm/envs.py

@@ -35,6 +35,7 @@
     VLLM_PP_LAYER_PARTITION: Optional[str] = None
     VLLM_CPU_KVCACHE_SPACE: int = 0
     VLLM_CPU_OMP_THREADS_BIND: str = ""
+    VLLM_OPENVINO_DEVICE: str = "CPU"
     VLLM_OPENVINO_KVCACHE_SPACE: int = 0
     VLLM_OPENVINO_CPU_KV_CACHE_PRECISION: Optional[str] = None
     VLLM_OPENVINO_ENABLE_QUANTIZED_WEIGHTS: bool = False
@@ -302,6 +303,11 @@ def get_default_config_root():
     "VLLM_CPU_OMP_THREADS_BIND":
     lambda: os.getenv("VLLM_CPU_OMP_THREADS_BIND", "all"),
 
+    # OpenVINO device selection
+    # default is CPU
+    "VLLM_OPENVINO_DEVICE":
+    lambda: os.getenv("VLLM_OPENVINO_DEVICE", "CPU").upper(),
+
     # OpenVINO key-value cache space
     # default is 4GB
     "VLLM_OPENVINO_KVCACHE_SPACE":

vllm/executor/openvino_executor.py

@@ -17,15 +17,28 @@
 logger = init_logger(__name__)
 
 
+def is_openvino_cpu() -> bool:
+    return "CPU" in envs.VLLM_OPENVINO_DEVICE
+
+
+def is_openvino_gpu() -> bool:
+    return "GPU" in envs.VLLM_OPENVINO_DEVICE
+
+
 class OpenVINOExecutor(ExecutorBase):
 
     uses_ray: bool = False
 
     def _init_executor(self) -> None:
         assert self.device_config.device_type == "openvino"
         assert self.lora_config is None, "OpenVINO backend doesn't support LoRA"
+        assert is_openvino_cpu() or is_openvino_gpu(), \
+            "OpenVINO backend supports only CPU and GPU devices"
+
+        self.ov_core = ov.Core()
         self.model_config = _verify_and_get_model_config(self.model_config)
-        self.cache_config = _verify_and_get_cache_config(self.cache_config)
+        self.cache_config = _verify_and_get_cache_config(
+            self.ov_core, self.cache_config)
 
         # Instantiate the worker and load the model to CPU.
         self._init_worker()
@@ -40,6 +53,7 @@ def _init_worker(self):
         distributed_init_method = get_distributed_init_method(
             get_ip(), get_open_port())
         self.driver_worker = OpenVINOWorker(
+            ov_core=self.ov_core,
             model_config=self.model_config,
             parallel_config=self.parallel_config,
             scheduler_config=self.scheduler_config,
@@ -68,10 +82,13 @@ def initialize_cache(self, num_gpu_blocks: int,
         # NOTE: We log here to avoid multiple logs when number of workers is
         # greater than one. We could log in the engine, but not all executors
         # have GPUs.
-        # NOTE: `cpu block` for OpenVINO backend is located on CPU memory but is
-        # referred as `gpu block`. Because we want to reuse the existing block
-        # management procedure.
-        logger.info("# CPU blocks: %d", num_gpu_blocks)
+        # NOTE: In case of a CPU device, `cpu block` for OpenVINO backend
+        # is located on CPU memory but is referred as `gpu block`.
+        # Because we want to reuse the existing block management procedure.
+        device_blocks = num_gpu_blocks
+        swap_blocks = num_cpu_blocks
+        logger.info("OpenVINO %s: # device blocks: %d; # swap blocks: %d",
+                    envs.VLLM_OPENVINO_DEVICE, device_blocks, swap_blocks)
         self.driver_worker.initialize_cache(num_gpu_blocks, num_cpu_blocks)
 
     def execute_model(
@@ -143,29 +160,45 @@ def _verify_and_get_model_config(config: ModelConfig) -> ModelConfig:
     return config
 
 
-def _verify_and_get_cache_config(config: CacheConfig) -> CacheConfig:
+def _verify_and_get_cache_config(ov_core: ov.Core,
+                                 config: CacheConfig) -> CacheConfig:
     if envs.VLLM_OPENVINO_CPU_KV_CACHE_PRECISION == "u8":
-        logger.info("KV cache type is overried to u8 via "
-                    "VLLM_OPENVINO_CPU_KV_CACHE_PRECISION env var.")
-        config.cache_dtype = ov.Type.u8
+        if not is_openvino_cpu():
+            logger.info("VLLM_OPENVINO_CPU_KV_CACHE_PRECISION is"
+                        "ignored for GPU, f16 data type will be used.")
+            config.cache_dtype = ov.Type.f16
+        else:
+            logger.info("KV cache type is overridden to u8 via "
+                        "VLLM_OPENVINO_CPU_KV_CACHE_PRECISION env var.")
+            config.cache_dtype = ov.Type.u8
     else:
-        core = ov.Core()
-        inference_precision = core.get_property("CPU",
-                                                hints.inference_precision)
-        if inference_precision == ov.Type.bf16:
-            config.cache_dtype = ov.Type.bf16
+        if is_openvino_cpu():
+            ov_device = envs.VLLM_OPENVINO_DEVICE
+            inference_precision = ov_core.get_property(
+                ov_device, hints.inference_precision)
+            if inference_precision == ov.Type.bf16:
+                config.cache_dtype = ov.Type.bf16
+            else:
+                config.cache_dtype = ov.Type.f16
         else:
             config.cache_dtype = ov.Type.f16
 
-    if config.block_size != 32:
-        logger.info(
-            f"OpenVINO optimal block size is 32, overriding currently set {config.block_size}"  # noqa: G004, E501
-        )
-        config.block_size = 32
+    if is_openvino_cpu():
+        if config.block_size != 32:
+            logger.info(
+                f"OpenVINO CPU optimal block size is 32, overriding currently set {config.block_size}"  # noqa: G004, E501
+            )
+            config.block_size = 32
+    else:
+        if config.block_size != 16:
+            logger.info(
+                f"OpenVINO GPU optimal block size is 16, overriding currently set {config.block_size}"  # noqa: G004, E501
+            )
+            config.block_size = 16
 
     kv_cache_space = envs.VLLM_OPENVINO_KVCACHE_SPACE
     if kv_cache_space >= 0:
-        if kv_cache_space == 0:
+        if kv_cache_space == 0 and is_openvino_cpu():
             config.openvino_kvcache_space_bytes = 4 * GiB_bytes  # type: ignore
             logger.warning(
                 "Environment variable VLLM_OPENVINO_KVCACHE_SPACE (GB) "

vllm/model_executor/model_loader/openvino.py

@@ -12,6 +12,7 @@
 import vllm.envs as envs
 from vllm.attention.backends.openvino import OpenVINOAttentionMetadata
 from vllm.config import DeviceConfig, ModelConfig
+from vllm.executor.openvino_executor import is_openvino_cpu
 from vllm.logger import init_logger
 from vllm.model_executor.layers.logits_processor import (LogitsProcessor,
                                                          _prune_hidden_states)
@@ -51,25 +52,15 @@ def _modify_cache_parameters(model: ov.Model, kv_cache_dtype: ov.Type,
         shape = parameter.get_partial_shape()
         # use real block size if available, just a placeholder
         # to provide the expected rank
-        x_size = 1
         num_blocks = ov.Dimension()
         block_size = ov.Dimension()
         head_size = ov.Dimension()
-        # TODO: Negotiate required layout with plugins (CPU is ~OK, GPU is TBD),
-        # pass more parameters to this function to set more static dimensions
         if input_name.startswith("key_cache."):
             cpu_shape = [num_blocks, shape[1], block_size, head_size]
-            gpu_shape = [
-                num_blocks,
-                shape[1],
-                shape[2].get_length() //
-                x_size if shape[2].is_static else ov.Dimension(),
-                block_size,
-                x_size,
-            ]
+            gpu_shape = [num_blocks, shape[1], shape[2], block_size]
         elif input_name.startswith("value_cache."):
             cpu_shape = [num_blocks, shape[1], block_size, head_size]
-            gpu_shape = [num_blocks, shape[1], shape[2], block_size]
+            gpu_shape = [num_blocks, shape[1], block_size, shape[2]]
         else:
             continue
         parameter.set_partial_shape(
@@ -108,6 +99,7 @@ class OpenVINOCasualLM(nn.Module):
 
     def __init__(
         self,
+        ov_core: ov.Core,
         model_config: ModelConfig,
         device_config: DeviceConfig,
         kv_cache_dtype: ov.Type,
@@ -141,12 +133,12 @@ def __init__(
             trust_remote_code=model_config.trust_remote_code,
         )
 
+        ov_device = envs.VLLM_OPENVINO_DEVICE
         paged_attention_transformation(pt_model.model)
         _modify_cache_parameters(pt_model.model, kv_cache_dtype,
-                                 device_config.device.type == "cpu")
+                                 is_openvino_cpu())
 
-        core = ov.Core()
-        ov_compiled = core.compile_model(pt_model.model, "CPU")
+        ov_compiled = ov_core.compile_model(pt_model.model, ov_device)
         self.ov_request = ov_compiled.create_infer_request()
 
     def forward(
@@ -199,11 +191,13 @@ def get_model(
     **kwargs,
 ) -> torch.nn.Module:
     lora_config = kwargs.get("lora_config", None)
+    ov_core = kwargs.get("ov_core")
     if lora_config:
         raise ValueError(
             "OpenVINO modeling does not support LoRA, "
             "but LoRA is enabled. Support for this model may "
             "be added in the future. If this is important to you, "
             "please open an issue on github.")
 
-    return OpenVINOCasualLM(model_config, device_config, kv_cache_dtype)
+    return OpenVINOCasualLM(ov_core, model_config, device_config,
+                            kv_cache_dtype)

vllm/worker/openvino_model_runner.py

@@ -42,6 +42,7 @@ class OpenVINOModelRunner:
 
     def __init__(
         self,
+        ov_core: ov.Core,
         model_config: ModelConfig,
         parallel_config: ParallelConfig,
         scheduler_config: SchedulerConfig,
@@ -55,6 +56,7 @@ def __init__(
         *args,
         **kwargs,
     ):
+        self.ov_core = ov_core
         self.model_config = model_config
         self.parallel_config = parallel_config
         self.scheduler_config = scheduler_config
@@ -89,11 +91,10 @@ def __init__(
         self.model: nn.Module  # Set after init_Model
 
     def load_model(self) -> None:
-        self.model = get_model(
-            model_config=self.model_config,
-            device_config=self.device_config,
-            kv_cache_dtype=self.kv_cache_dtype,
-        )
+        self.model = get_model(model_config=self.model_config,
+                               device_config=self.device_config,
+                               kv_cache_dtype=self.kv_cache_dtype,
+                               ov_core=self.ov_core)
 
     def _prepare_model_input(
         self,