[源码解析] NVIDIA HugeCTR，GPU版本参数服务器--- (4)

[源码解析] NVIDIA HugeCTR，GPU版本参数服务器--- (4)

0x00 摘要

在这个系列中，我们介绍了 HugeCTR，这是一个面向行业的推荐系统训练框架，针对具有模型并行嵌入和数据并行密集网络的大规模 CTR 模型进行了优化。

本文主要介绍流水线的前两级，最后一级将会独立成文。其中借鉴了HugeCTR源码阅读 这篇大作，特此感谢。

本系列其他文章如下：

[源码解析] NVIDIA HugeCTR，GPU 版本参数服务器 --(1)

[源码解析] NVIDIA HugeCTR，GPU版本参数服务器--- (2)

[源码解析] NVIDIA HugeCTR，GPU版本参数服务器---(3)

0x01 总体流程

由于高效的数据交换和三级流水线，HugeCTR的可扩展性和活跃GPU的数量都有所增加。此流水线包括三级：

• 从文件读取数据。
• 从主机到设备的数据传输（节点间和节点内）。
• 利用GPU计算。

的数据读取重叠，并训练GPU。下图显示了HugeCTR的可扩展性，批量大小为16384，在DGX1服务器上有七层。

0x02 DataReader

DataReader 被用来把数据从数据集拷贝到嵌入层。其是流水线的入口，包括了流水线的前面两步骤：读取文件和拷贝到GPU。

Figure 5. HugeCTR training pipeline with its data reader.

2.1 定义

为了分析需要，我们只给出成员变量，方法我们会在使用时候具体介绍。

从动态角度看，成员变量之中重要的是以下两个：

• worker_group ：工作线程组，负责把数据从dataset文件读取到内存之中，这个可以认为是流水线的第一级。之前的版本之中有一个HeapEx数据结构用来做中间缓存，目前这个数据结构已经移除。
• data_collector_ ：拥有一个线程，负责把数据拷贝到GPU之中。这个可以认为是流水线的第二级。

从静态角度看，主要是以下三个buffer：

• std::vector> thread_buffers_：线程内部使用的buffer。
• std::shared_ptr broadcast_buffer_：用来后续和collector交互，collector 把它作为中间buffer。
• std::shared_ptr output_：reader的输出，训练最后读取的是这里。

以上三个buffer的数据流动是：ThreadBuffer --> BroadcastBuffer ---> DataReaderOutput。

从资源角度看，则是：

• std::shared_ptr resource_manager_ ：这是 Session 的成员变量，在DataReader构造函数之中传递进来的。
• const std::vector params_ ：这是依据配置文件整理出来的sparse参数元信息。

/**
 * @brief Data reading controller.
 *
 * Control the data reading from data set to embedding.
 * An instance of DataReader will maintain independent
 * threads for data reading (IDataReaderWorker)
 * from dataset to heap. Meanwhile one independent
 * thread consumes the data (DataCollector),
 * and copy the data to GPU buffer.
 */
template 
class DataReader : public IDataReader {
 private:
  std::vector> thread_buffers_;  // gpu_id -> thread_idx
  std::shared_ptr broadcast_buffer_;
  std::shared_ptr output_;

  std::shared_ptr worker_group_;
  std::shared_ptr> data_collector_; /**< pointer of DataCollector */

  /* Each gpu will have several csr output for different embedding */
  const std::vector params_;
  std::shared_ptr resource_manager_; /**< gpu resource used in this data reader*/
  const size_t batchsize_;                            /**< batch size */
  const size_t label_dim_; /**< dimention of label e.g. 1 for BinaryCrossEntropy */
  const size_t dense_dim_; /**< dimention of dense */
  long long current_batchsize_;

  bool repeat_;
  std::string file_name_;
  SourceType_t source_type_;
}

2.2 构建

对DataReader的构建分为两部分：

• 在构造函数之中会：
  • 对各种buffer进行配置。
  • 对构建DataCollector。
• 在create_datareader之中会分别处理 train_data_reader和 evaluate_data_reader，也就是用于训练和评估的两个reader。然后会为他们建立workgroup。

我们先省略对构造函数的分析，因为其牵扯到一系列数据结构。等介绍完数据结构之后，再进行论述。

2.3 DataReaderSparseParam

2.3.1 定义

DataReaderSparseParam 是依据配置得到的Sparse参数的元信息，其主要成员变量如下：

• sparse_name是其后续层引用的稀疏输入张量的名称。没有默认值，应由用户指定。

• nnz_per_slot是每个插槽的指定sparse输入的最大特征数。

  • 'nnz_per_slot'可以是'int'，即每个slot的平均nnz，因此每个实例的最大功能数应该是'nnz_per_slot*slot_num'。
  • 或者可以使用List[int]初始化'nnz_per_slot'，则每个样本的最大特征数应为'sum(nnz_per_slot)'，在这种情况下，数组'nnz_per_slot'的长度应与'slot_num'相同。

• 'is_fixed_length'用于标识所有样本中每个插槽的categorical inputs是否具有相同的长度。如果不同的样本对于每个插槽具有相同数量的特征，则用户可以设置“is_fixed_length=True”，Hugetr可以使用此信息来减少数据传输时间。

• slot_num指定用于数据集中此稀疏输入的插槽数。

  • 注意：如果指定了多个'DataReaderSparseParam'，则任何一对'DataReaderSparseParam'之间都不应有重叠。比如，在[wdl样本]（../samples/wdl/wdl.py）中，我们总共有27个插槽；我们将第一个插槽指定为"wide_data"，将接下来的26个插槽指定为"deep_data"。

struct DataReaderSparseParam {
  std::string top_name;
  std::vector nnz_per_slot;
  bool is_fixed_length;
  int slot_num;

  DataReaderSparse_t type;
  int max_feature_num;
  int max_nnz;

  DataReaderSparseParam() {}
  DataReaderSparseParam(const std::string& top_name_, const std::vector& nnz_per_slot_,
                        bool is_fixed_length_, int slot_num_)
      : top_name(top_name_),
        nnz_per_slot(nnz_per_slot_),
        is_fixed_length(is_fixed_length_),
        slot_num(slot_num_),
        type(DataReaderSparse_t::Distributed) {
    max_feature_num = std::accumulate(nnz_per_slot.begin(), nnz_per_slot.end(), 0);
    max_nnz = *std::max_element(nnz_per_slot.begin(), nnz_per_slot.end());
  }

  DataReaderSparseParam(const std::string& top_name_, const int nnz_per_slot_,
                        bool is_fixed_length_, int slot_num_)
      : top_name(top_name_),
        nnz_per_slot(slot_num_, nnz_per_slot_),
        is_fixed_length(is_fixed_length_),
        slot_num(slot_num_),
        type(DataReaderSparse_t::Distributed) {
    max_feature_num = std::accumulate(nnz_per_slot.begin(), nnz_per_slot.end(), 0);
    max_nnz = *std::max_element(nnz_per_slot.begin(), nnz_per_slot.end());
  }
};

2.3.2 使用

之前提到了Parser是解析配置文件，HugeCTR 也支持代码设置，比如下面就设定了两个DataReaderSparseParam，也有对应的DistributedSlotSparseEmbeddingHash。

model = hugectr.Model(solver, reader, optimizer)
model.add(hugectr.Input(label_dim = 1, label_name = "label",
                        dense_dim = 13, dense_name = "dense",
                        data_reader_sparse_param_array = 
                        [hugectr.DataReaderSparseParam("wide_data", 30, True, 1),
                        hugectr.DataReaderSparseParam("deep_data", 2, False, 26)]))
model.add(hugectr.SparseEmbedding(embedding_type = hugectr.Embedding_t.DistributedSlotSparseEmbeddingHash, 
                            workspace_size_per_gpu_in_mb = 23,
                            embedding_vec_size = 1,
                            combiner = "sum",
                            sparse_embedding_name = "sparse_embedding2",
                            bottom_name = "wide_data",
                            optimizer = optimizer))
model.add(hugectr.SparseEmbedding(embedding_type = hugectr.Embedding_t.DistributedSlotSparseEmbeddingHash, 
                            workspace_size_per_gpu_in_mb = 358,
                            embedding_vec_size = 16,
                            combiner = "sum",
                            sparse_embedding_name = "sparse_embedding1",
                            bottom_name = "deep_data",
                            optimizer = optimizer))

0x03 DataReader Buffer 机制

我们接下来看看 DataReader 的若干Buffer，依赖于这些buffer，HugeCTR实现了流水线的前两级。

3.1 比对

我们首先要做一个历史对比，看看这部分代码的发展脉络。我们先看看3.1版本的代码。DataReader 我们选取了部分成员变量。3.1 版本之前使用了一个heap进行操作，即下面的csr_heap_。

class DataReader : public IDataReader {
  std::shared_ptr>> csr_heap_; /**< heap to cache the data set */
  Tensors2 label_tensors_;                       /**< Label tensors for the usage of loss */
  std::vector dense_tensors_;               /**< Dense tensors for the usage of loss */
  /* Each gpu will have several csr output for different embedding */
  Tensors2 csr_buffers_; /**< csr_buffers contains row_offset_tensor and value_tensors */
  Tensors2 row_offsets_tensors_; /**< row offset tensors*/
  Tensors2 value_tensors_;       /**< value tensors */
  std::vector> nnz_array_;

  const size_t label_dim_; /**< dimention of label e.g. 1 for BinaryCrossEntropy */
  const size_t dense_dim_; /**< dimention of dense */
}

我们再看看3.2.1版本的代码，也选取了部分成员变量。

template 
class DataReader : public IDataReader {
  std::vector> thread_buffers_;  // gpu_id -> thread_idx
  std::shared_ptr broadcast_buffer_;
  std::shared_ptr output_;

  const size_t label_dim_; /**< dimention of label e.g. 1 for BinaryCrossEntropy */
  const size_t dense_dim_; /**< dimention of dense */
}

3.2.1 这里是：

• 把 label_tensors_, dense_tensors_ 移动到 AsyncReader。
• 把 csr_heap_ 用 thread_buffers_，broadcast_buffer_，output_ 等进行替代。
• 把 row_offsets_tensors_，value_tensors_，nnz_array_ 等等用 ThreadBuffer，BroadcastBuffer，DataReaderOutput 之中的 SparseTensorBag 来包括，统一管理 CSR。

3.2 Buffer 相关类

我们依据上面的历史版本比对来看看。

• 在之前版本（比如3.1）之中，存在一个 HeapEX 类，其实现了 CPU 到 GPU 之间的一个数据缓存功能。
• 在最新版本之中，改为一系列 buffer 相关类，比如 ThreadBuffer 和 BroadcastBuffer，其状态都是由 BufferState 实现的。

enum class BufferState : int { FileEOF, Reading, ReadyForRead, Writing, ReadyForWrite };

以下是三个buffer的定义。

struct ThreadBuffer {
  std::vector device_sparse_buffers;  // same number as embedding number
  std::vector is_fixed_length;          // same number as embedding number
  TensorBag2 device_dense_buffers;
  std::atomic state;
  long long current_batch_size;
  int batch_size;
  size_t param_num;
  int label_dim;
  int dense_dim;
  int batch_size_start_idx;  // dense buffer
  int batch_size_end_idx;
};

struct BroadcastBuffer {
  std::vector
      sparse_buffers;  // same number as (embedding number * local device number)
  std::vector is_fixed_length;        // same number as embedding number
  std::vector dense_tensors;             // same number as local device number
  std::vector finish_broadcast_events;  // same number as local device number
  std::atomic state;
  long long current_batch_size;
  size_t param_num;
};

struct DataReaderOutput {
  std::map> sparse_tensors_map;
  std::vector sparse_name_vec;
  std::vector label_tensors;
  std::vector dense_tensors;
  bool use_mixed_precision;
  int label_dense_dim;
};

以上这些类，对应了 DataReader 的以下成员变量。

class DataReader : public IDataReader {
 private:
  std::vector> thread_buffers_;  // gpu_id -> thread_idx
  std::shared_ptr broadcast_buffer_;
  std::shared_ptr output_;
}

接下来，我们就一一分析。

3.3 DataReader构造

前面跳过了 DataReader 的构造函数，接下来我们接下来对构造函数进行分析，其主要功能就是为三种buffer来预留空间，分配内存，最后构建了collector。

DataReader(int batchsize, size_t label_dim, int dense_dim,
           std::vector ¶ms,
           const std::shared_ptr &resource_manager, bool repeat, int num_threads,
           bool use_mixed_precision)
    : broadcast_buffer_(new BroadcastBuffer()),
      output_(new DataReaderOutput()),
      params_(params),
      resource_manager_(resource_manager),
      batchsize_(batchsize),
      label_dim_(label_dim),
      dense_dim_(dense_dim),
      repeat_(repeat) {
  size_t local_gpu_count = resource_manager_->get_local_gpu_count();
  size_t total_gpu_count = resource_manager_->get_global_gpu_count();

  // batchsize_ is a multiple of total_gpu_count
  size_t batch_size_per_gpu = batchsize_ / total_gpu_count;
        
  // 1. 生成了一个临时变量buffs，用来具体分配内存，里面是若干 CudaAllocator，每个CudaAllocator对应了i个GPU 
  std::vector>> buffs;
  // 先预留部分内存空间      
  buffs.reserve(local_gpu_count);
  // 为每个GPU初始化一个GeneralBuffer2   
  for (size_t i = 0; i < local_gpu_count; ++i) {
    buffs.push_back(GeneralBuffer2::create());
  }

  // 2.预留buffer 
  // 处理 thread_buffers_     
  thread_buffers_.reserve(num_threads);
  for (int i = 0; i < num_threads; ++i) { 
    // a worker may maintain multiple buffers on device i % local_gpu_count
    auto local_gpu = resource_manager_->get_local_gpu(i % local_gpu_count);
    CudaCPUDeviceContext context(local_gpu->get_device_id());
    auto &buff = buffs[i % local_gpu_count]; // 找到对应GPU对应的CudaAllocator，进行分配
    std::shared_ptr current_thread_buffer = std::make_shared();
    thread_buffers_.push_back(current_thread_buffer);

    current_thread_buffer->device_sparse_buffers.reserve(params.size());
    current_thread_buffer->is_fixed_length.reserve(params.size()); // vector的reserve
    for (size_t param_id = 0; param_id < params.size(); ++param_id) {
      auto ¶m = params_[param_id];
      SparseTensor temp_sparse_tensor;
      // 预留内存
      buff->reserve({(size_t)batchsize, (size_t)param.max_feature_num}, param.slot_num,
                    &temp_sparse_tensor);
      current_thread_buffer->device_sparse_buffers.push_back(temp_sparse_tensor.shrink());
      current_thread_buffer->is_fixed_length.push_back(param.is_fixed_length);
    }
    Tensor2 temp_dense_tensor;
    // 预留内存
    buff->reserve({batch_size_per_gpu * local_gpu_count, label_dim + dense_dim},
                  &temp_dense_tensor);
    current_thread_buffer->device_dense_buffers = temp_dense_tensor.shrink();
    current_thread_buffer->state.store(BufferState::ReadyForWrite);
    current_thread_buffer->current_batch_size = 0;
    current_thread_buffer->batch_size = batchsize;
    current_thread_buffer->param_num = params.size();
    current_thread_buffer->label_dim = label_dim;
    current_thread_buffer->dense_dim = dense_dim;
    current_thread_buffer->batch_size_start_idx =
        batch_size_per_gpu * resource_manager_->get_gpu_global_id_from_local_id(0);
    current_thread_buffer->batch_size_end_idx =
        current_thread_buffer->batch_size_start_idx + batch_size_per_gpu * local_gpu_count;
  }

  // 处理 broadcast buffer，注意这里的reserve是 vector数据结构的方法，不是预留内存      
  broadcast_buffer_->sparse_buffers.reserve(local_gpu_count * params.size());
  broadcast_buffer_->is_fixed_length.reserve(local_gpu_count * params.size());
  broadcast_buffer_->dense_tensors.reserve(local_gpu_count);
  broadcast_buffer_->finish_broadcast_events.resize(local_gpu_count);
  broadcast_buffer_->state.store(BufferState::ReadyForWrite);
  broadcast_buffer_->current_batch_size = 0;
  broadcast_buffer_->param_num = params.size();
        
  // 处理 output buffer，注意这里的reserve是 vector数据结构的方法，不是预留内存
  output_->dense_tensors.reserve(local_gpu_count);
  output_->label_tensors.reserve(local_gpu_count);
  output_->use_mixed_precision = use_mixed_precision;
  output_->label_dense_dim = label_dim + dense_dim;
  // 预留sparse tensor，注意这里的reserve是 vector数据结构的方法，不是预留内存      
  for (size_t param_id = 0; param_id < params.size(); ++param_id) {
    auto ¶m = params_[param_id];
    output_->sparse_tensors_map[param.top_name].reserve(local_gpu_count);
    output_->sparse_name_vec.push_back(param.top_name);
  }

  // 遍历本地的 GPU       
  for (size_t local_id = 0; local_id < local_gpu_count; ++local_id) {
    // 还是需要针对每一个GPU，找到对应的CudaAllocator进行分配
    auto local_gpu = resource_manager_->get_local_gpu(local_id);
    CudaDeviceContext ctx(local_gpu->get_device_id());
    auto &buff = buffs[local_id];

    for (size_t param_id = 0; param_id < params.size(); ++param_id) {
      auto ¶m = params_[param_id];
      SparseTensor temp_sparse_tensor;
      // 给broadcast_buffer_分配内存
      buff->reserve({(size_t)batchsize, (size_t)param.max_feature_num}, param.slot_num,
                    &temp_sparse_tensor);
      broadcast_buffer_->sparse_buffers.push_back(temp_sparse_tensor.shrink());
      broadcast_buffer_->is_fixed_length.push_back(param.is_fixed_length);
    }
    Tensor2 temp_dense_tensor;
    buff->reserve({batch_size_per_gpu, label_dim + dense_dim}, &temp_dense_tensor);
    broadcast_buffer_->dense_tensors.push_back(temp_dense_tensor.shrink());

    CK_CUDA_THROW_(cudaEventCreateWithFlags(&broadcast_buffer_->finish_broadcast_events[local_id],
                                            cudaEventDisableTiming));

    for (size_t param_id = 0; param_id < params.size(); ++param_id) {
      auto ¶m = params_[param_id];
      SparseTensor temp_sparse_tensor;
      // 预留内存
      buff->reserve({(size_t)batchsize, (size_t)param.max_feature_num}, param.slot_num,
                    &temp_sparse_tensor);
      output_->sparse_tensors_map[param.top_name].push_back(temp_sparse_tensor.shrink());
    }

    Tensor2 label_tensor;
    // 预留内存
    buff->reserve({batch_size_per_gpu, label_dim}, &label_tensor);
    output_->label_tensors.push_back(label_tensor.shrink());

    if (use_mixed_precision) {
      Tensor2<__half> dense_tensor;
      // 预留内存
      buff->reserve({(size_t)batch_size_per_gpu, (size_t)dense_dim}, &dense_tensor);
      output_->dense_tensors.push_back(dense_tensor.shrink());
    } else {
      Tensor2 dense_tensor;
      // 预留内存
      buff->reserve({(size_t)batch_size_per_gpu, (size_t)dense_dim}, &dense_tensor);
      output_->dense_tensors.push_back(dense_tensor.shrink());
    }

    buff->allocate(); // 3. 分配内存
  }

  // 4. 构建DataCollector     
  data_collector_ = std::make_shared>(thread_buffers_, broadcast_buffer_,
                                                             output_, resource_manager);
  return;
}

我们接下来会仔细分一下构造代码之中的各个部分。

3.3.1 辅助 GeneralBuffer2

首先我们分析上面代码之中buffs部分，这个变量作用就是统一分配内存。

  // 1. 生成了一个临时变量buffs    
  std::vector>> buffs;
  // 先预留部分容量大小     
  buffs.reserve(local_gpu_count);
  // 为每个GPU初始化一个GeneralBuffer2   
  for (size_t i = 0; i < local_gpu_count; ++i) {
    buffs.push_back(GeneralBuffer2::create());
  }

3.3.2 ThreadBuffer

然后我们看看处理 thread_buffers_ 部分，这里是为线程buffer进行处理。我们首先获取ThreadBuffer类定义如下，后面分析时候可以比对。

struct ThreadBuffer {
  std::vector device_sparse_buffers;  // same number as embedding number
  std::vector is_fixed_length;          // same number as embedding number
  TensorBag2 device_dense_buffers;
  std::atomic state;
  long long current_batch_size;
  int batch_size;
  size_t param_num;
  int label_dim;
  int dense_dim;
  int batch_size_start_idx;  // dense buffer
  int batch_size_end_idx;
};

其次，具体构建函数中的逻辑如下：

• 首先，对于 thread_buffers_ 这个vector，会拓展 vector 容量到线程数大小。
• 拿到本线程（或者说是本GPU）在buffs之中对应的buffer，赋值到 buff。
• 对于每一个线程，会生成一个ThreadBuffer，命名为current_thread_buffer，放入到 thread_buffers_ 之中。
• 对于每一个 ThreadBuffer，预留 ThreadBuffer 的device_sparse_buffers 和 is_fixed_length 这两个 vector 的容量大小。
• 遍历sparse参数，对于每一个参数，会建立一个临时张量，并且通过 buff 预留内存（CPU或者GPU），然后把此临时张量放入device_sparse_buffers。
• 建立一个针对dense的张量，并且通过 buff 预留张量内存，把临时张量放入device_dense_buffers。
• 设置current_thread_buffer 状态。
• 设置 current_thread_buffer 其他信息。

  // 处理 thread_buffers_，会拓展 vector 容量到线程数大小 
  thread_buffers_.reserve(num_threads);
  for (int i = 0; i < num_threads; ++i) {  // 遍历线程
    // a worker may maintain multiple buffers on device i % local_gpu_count
    auto local_gpu = resource_manager_->get_local_gpu(i % local_gpu_count);
    CudaCPUDeviceContext context(local_gpu->get_device_id());
    auto &buff = buffs[i % local_gpu_count]; // 拿到本线程（或者说是本GPU）在buffs之中对应的buffer
    // 生成一个ThreadBuffer，存入到thread_buffers_
    std::shared_ptr current_thread_buffer = std::make_shared();
    thread_buffers_.push_back(current_thread_buffer);

    // 预留 ThreadBuffer 的device_sparse_buffers 和 is_fixed_length 这两个 vector 的容量大小
    current_thread_buffer->device_sparse_buffers.reserve(params.size());
    current_thread_buffer->is_fixed_length.reserve(params.size());
    
    // 遍历参数
    for (size_t param_id = 0; param_id < params.size(); ++param_id) {
      auto ¶m = params_[param_id];
      SparseTensor temp_sparse_tensor;
      // 建立一个临时张量，并且预留内存（CPU或者GPU）
      buff->reserve({(size_t)batchsize, (size_t)param.max_feature_num}, param.slot_num,
                    &temp_sparse_tensor);
      // 把张量放入device_sparse_buffers
      current_thread_buffer->device_sparse_buffers.push_back(temp_sparse_tensor.shrink());
      current_thread_buffer->is_fixed_length.push_back(param.is_fixed_length);
    }
    
    // 建立一个针对dense的张量
    Tensor2 temp_dense_tensor;
    // 预留张量内存
    buff->reserve({batch_size_per_gpu * local_gpu_count, label_dim + dense_dim},
                  &temp_dense_tensor);
    // 把临时张量放入device_dense_buffers
    current_thread_buffer->device_dense_buffers = temp_dense_tensor.shrink();
    // 设置状态
    current_thread_buffer->state.store(BufferState::ReadyForWrite);
    // 设置其他信息
    current_thread_buffer->current_batch_size = 0;
    current_thread_buffer->batch_size = batchsize;
    current_thread_buffer->param_num = params.size();
    current_thread_buffer->label_dim = label_dim;
    current_thread_buffer->dense_dim = dense_dim;
    current_thread_buffer->batch_size_start_idx =
        batch_size_per_gpu * resource_manager_->get_gpu_global_id_from_local_id(0);
    current_thread_buffer->batch_size_end_idx =
        current_thread_buffer->batch_size_start_idx + batch_size_per_gpu * local_gpu_count;
  }

此时如下，注意，DataReader 包括多个 ThreadBuffer。

3.3.3 BroadcastBuffer

接下来看看如何构建BroadcastBuffer。

BroadcastBuffer定义如下：

struct BroadcastBuffer {
  std::vector
      sparse_buffers;  // same number as (embedding number * local device number)
  std::vector is_fixed_length;        // same number as embedding number
  std::vector dense_tensors;             // same number as local device number
  std::vector finish_broadcast_events;  // same number as local device number
  std::atomic state;
  long long current_batch_size;
  size_t param_num;
};

按照构建代码来说，这里只是做了一些预留和设置，没有涉及内存，内存在后续会统一处理。

  // 处理 broadcast buffer      
  // 预留vector的容量
  broadcast_buffer_->sparse_buffers.reserve(local_gpu_count * params.size());
  // 预留vector的容量
  broadcast_buffer_->is_fixed_length.reserve(local_gpu_count * params.size());
  // 预留vector的容量
  broadcast_buffer_->dense_tensors.reserve(local_gpu_count);
  broadcast_buffer_->finish_broadcast_events.resize(local_gpu_count);
  // 设置状态
  broadcast_buffer_->state.store(BufferState::ReadyForWrite);
  broadcast_buffer_->current_batch_size = 0;
  broadcast_buffer_->param_num = params.size();

3.3.4 DataReaderOutput

我们接着看看如何构建DataReaderOutput。

struct DataReaderOutput {
  std::map> sparse_tensors_map;
  std::vector sparse_name_vec;
  std::vector label_tensors;
  std::vector dense_tensors;
  bool use_mixed_precision;
  int label_dense_dim;
};

按照构建代码来说，这里只是做了一些预留和设置，没有涉及内存，内存在后续会统一处理。

output_->dense_tensors.reserve(local_gpu_count); // 预留vector的容量
output_->label_tensors.reserve(local_gpu_count); // 预留vector的容量
output_->use_mixed_precision = use_mixed_precision;
output_->label_dense_dim = label_dim + dense_dim;
for (size_t param_id = 0; param_id < params.size(); ++param_id) {
  auto ¶m = params_[param_id];

  output_->sparse_tensors_map[param.top_name].reserve(local_gpu_count);
  output_->sparse_name_vec.push_back(param.top_name);
}

3.3.5 预留和分配

这里会对 broadcast 和 output 进行预留，这里统一分配内存。

for (size_t local_id = 0; local_id < local_gpu_count; ++local_id) { // 遍历GPU
  auto local_gpu = resource_manager_->get_local_gpu(local_id);
  CudaDeviceContext ctx(local_gpu->get_device_id());
  auto &buff = buffs[local_id]; // 获取临时buffs之中对应某一个本地gpu的allocator

  for (size_t param_id = 0; param_id < params.size(); ++param_id) {
    auto ¶m = params_[param_id];
    SparseTensor temp_sparse_tensor;
    // 分配sparse内存
    buff->reserve({(size_t)batchsize, (size_t)param.max_feature_num}, param.slot_num,
                  &temp_sparse_tensor);
    // 赋值到broadcast 之上
    broadcast_buffer_->sparse_buffers.push_back(temp_sparse_tensor.shrink());
    broadcast_buffer_->is_fixed_length.push_back(param.is_fixed_length);
  }
  // 分配dense内存
  Tensor2 temp_dense_tensor;
  buff->reserve({batch_size_per_gpu, label_dim + dense_dim}, &temp_dense_tensor);
  // 赋值到broadcast 之上
  broadcast_buffer_->dense_tensors.push_back(temp_dense_tensor.shrink());

  CK_CUDA_THROW_(cudaEventCreateWithFlags(&broadcast_buffer_->finish_broadcast_events[local_id],
                                          cudaEventDisableTiming));

  for (size_t param_id = 0; param_id < params.size(); ++param_id) {
    auto ¶m = params_[param_id];

    // 分配sparse内存
    SparseTensor temp_sparse_tensor;
    buff->reserve({(size_t)batchsize, (size_t)param.max_feature_num}, param.slot_num,
                  &temp_sparse_tensor);
    // 赋值到output之上
    output_->sparse_tensors_map[param.top_name].push_back(temp_sparse_tensor.shrink());
  }

  // 分配label的内存
  Tensor2 label_tensor;
  buff->reserve({batch_size_per_gpu, label_dim}, &label_tensor);
  // 赋值到output之上
  output_->label_tensors.push_back(label_tensor.shrink());

  if (use_mixed_precision) {
    Tensor2<__half> dense_tensor;
    // 分配dense内存
    buff->reserve({(size_t)batch_size_per_gpu, (size_t)dense_dim}, &dense_tensor);
    // 赋值到output之上
    output_->dense_tensors.push_back(dense_tensor.shrink());
  } else {
    Tensor2 dense_tensor;
    // 分配dense内存
    buff->reserve({(size_t)batch_size_per_gpu, (size_t)dense_dim}, &dense_tensor);
    // 赋值到output之上
    output_->dense_tensors.push_back(dense_tensor.shrink());
  }

  buff->allocate(); // 统一分配
}

预留buffer的具体逻辑如下：

分配之后如下，需要注意的是，这里都是简化版本，没有体现出来多个本地GPU的状态。比如下面三个类的成员变量都会分配到多个本地GPU之上。

// embedding number 指的是本模型之中，DataReaderSparseParam 的个数，就是有几个 embedding 层
struct ThreadBuffer {
  std::vector device_sparse_buffers;  // same number as embedding number
  // device_sparse_buffers 会分配在多个本地GPU之上
  
struct BroadcastBuffer {
  std::vector
      sparse_buffers;  // same number as (embedding number * local device number)
  // sparse_buffers 也会分配在多个本地GPU之上

struct DataReaderOutput {
  std::map> sparse_tensors_map;
  // 每个 sparse_tensors_map[param.top_name] 都会分配在多个本地GPU之上
  // 比如 output_->sparse_tensors_map[param.top_name].reserve(local_gpu_count);

如下简化版本之中都只体现了一个GPU，这些buffer都是位于GPU之上。

现在 DataReader 有了一系列buffer，我们接下来看看如何使用。

0x04 DataReaderWorkerGroup

DataReaderWorkerGroup 负责具体读数据操作。

4.1 构建

在 create_datareader 之中，有如下代码建立 DataReaderWorkerGroup，分别对应了三种group。

    switch (format) {
      case DataReaderType_t::Norm: {
        train_data_reader->create_drwg_norm(source_data, check_type, start_right_now);
        evaluate_data_reader->create_drwg_norm(eval_source, check_type, start_right_now);
        break;
      }
      case DataReaderType_t::Raw: {
        train_data_reader->create_drwg_raw(source_data, num_samples, float_label_dense, true,
                                           false);
        evaluate_data_reader->create_drwg_raw(eval_source, eval_num_samples, float_label_dense,
                                              false, false);
        break;
      }
      case DataReaderType_t::Parquet: {
        train_data_reader->create_drwg_parquet(source_data, slot_offset, true);
        evaluate_data_reader->create_drwg_parquet(eval_source, slot_offset, true);
        break;
      }

我们用create_drwg_norm来继续分析，发现其构建了DataReaderWorkerGroupNorm。即，配置了 DataReader 之中的成员变量 worker_group_ 为一个 DataReaderWorkerGroupNorm。

注意，这里传入的是thread_buffers_，说明 DataReaderWorkerGroup 操作的就是DataReader 的 thread_buffers_。

void create_drwg_norm(std::string file_name, Check_t check_type,
                      bool start_reading_from_beginning = true) override {
  source_type_ = SourceType_t::FileList;
  worker_group_.reset(new DataReaderWorkerGroupNorm(
      thread_buffers_, resource_manager_, file_name, repeat_, check_type, params_,
      start_reading_from_beginning));
  file_name_ = file_name;
}

4.2 DataReaderWorkerGroup 定义

我们只看其成员变量，主要是 IDataReaderWorker，这就是具体读数据的wroker。

class DataReaderWorkerGroup {
  std::vector data_reader_threads_; /**< A vector of the pointers of data reader .*/
 protected:
  int data_reader_loop_flag_{0}; /**< p_loop_flag a flag to control the loop */
  DataReaderType_t data_reader_type_;
  std::vector>
      data_readers_; /**< A vector of DataReaderWorker' pointer.*/
  std::shared_ptr resource_manager_;
}

4.3 DataReaderWorkerGroupNorm

我们使用 DataReaderWorkerGroupNorm 来分析，其最重要的是构建 DataReaderWorker 时候，设定了每个DataReaderWorker 对应哪些GPU资源。

template 
class DataReaderWorkerGroupNorm : public DataReaderWorkerGroup {
  std::string file_list_; /**< file list of data set */

  std::shared_ptr create_source(size_t worker_id, size_t num_worker,
                                        const std::string &file_name, bool repeat) override {
    return std::make_shared(worker_id, num_worker, file_name, repeat);
  }

 public:
  // Ctor
  DataReaderWorkerGroupNorm(const std::vector> &output_buffers,
                            const std::shared_ptr &resource_manager_,
                            std::string file_list, bool repeat, Check_t check_type,
                            const std::vector ¶ms,
                            bool start_reading_from_beginning = true)
      : DataReaderWorkerGroup(start_reading_from_beginning, DataReaderType_t::Norm) {

    int num_threads = output_buffers.size();
    size_t local_gpu_count = resource_manager_->get_local_gpu_count();

    // create data reader workers
    int max_feature_num_per_sample = 0;
    for (auto ¶m : params) {
      max_feature_num_per_sample += param.max_feature_num;
    }

    set_resource_manager(resource_manager_);
    for (int i = 0; i < num_threads; i++) {
      std::shared_ptr data_reader(new DataReaderWorker(
          // 这里设定了每个 DataReaderWorker 对应的 GPU 资源
          i, num_threads, resource_manager_->get_local_gpu(i % local_gpu_count),
          &data_reader_loop_flag_, output_buffers[i], file_list, max_feature_num_per_sample, repeat,
          check_type, params));
      data_readers_.push_back(data_reader);
    }
    create_data_reader_threads(); // 建立了多个工作线程
  }
};

4.4 建立线程

create_data_reader_threads 建立了多个工作线程，设定了每个线程对应的 GPU 资源。

  /**
   * Create threads to run data reader workers
   */
  void create_data_reader_threads() {
    size_t local_gpu_count = resource_manager_->get_local_gpu_count();

    for (size_t i = 0; i < data_readers_.size(); ++i) {
      // 这里设定了每个线程对应的 GPU 资源
      auto local_gpu = resource_manager_->get_local_gpu(i % local_gpu_count);
      // 指定了线程主体函数
      data_reader_threads_.emplace_back(data_reader_thread_func_, data_readers_[i],
                                        &data_reader_loop_flag_, local_gpu->get_device_id());
    }
  }

4.5 线程主体函数

data_reader_thread_func_ 是工作线程的主体函数，里面设定了本线程的设备，然后调用了 IDataReaderWorker 完成读取数据。

/**
 * A helper function to read data from dataset to heap in a new thread.
 * @param data_reader a pointer of data_reader.
 * @param p_loop_flag a flag to control the loop,
          and break loop when IDataReaderWorker is destroyed.
 */
static void data_reader_thread_func_(const std::shared_ptr& data_reader,
                                     int* p_loop_flag, int device_id) {
  try {
    CudaCPUDeviceContext context(device_id); // 设定了本线程的设备

    while ((*p_loop_flag) == 0) {
      usleep(2);
    }

    while (*p_loop_flag) {
      data_reader->read_a_batch(); // 然后开始读取文件数据
    }
  } catch (const std::runtime_error& rt_err) {
    std::cerr << rt_err.what() << std::endl;
  }
}

所以，这里就设定了哪个样本应该放到哪个卡上，例如，下面4个线程，分别对应了 GPU 0 和 GPU 1。

4.6 DataReaderWorker

DataReaderWorker 是解析数据的业务模块。IDataReaderWorker 是 基类，其buffer_是关键，其指向了ThreadBuffer。

class IDataReaderWorker {
  std::shared_ptr source_; /**< source: can be file or network */

  int worker_id_;
  int worker_num_;
  std::shared_ptr gpu_resource_; // 这是本worker的GPU资源

  bool is_eof_;
  int *loop_flag_;

  std::shared_ptr buffer_;
  
  IDataReaderWorker(const int worker_id, const int worker_num,
                    const std::shared_ptr &gpu_resource, bool is_eof, int *loop_flag,
                    const std::shared_ptr &buff)
      : worker_id_(worker_id),
        worker_num_(worker_num),
        gpu_resource_(gpu_resource), // 设定GPU资源
        is_eof_(is_eof),
        loop_flag_(loop_flag),
        buffer_(buff) {}  
};

DataReaderWorker 具体定义如下：

template 
class DataReaderWorker : public IDataReaderWorker {
 private:
  DataSetHeader
      data_set_header_;  /**< the header of data set, which has main informations of a data file */
  size_t buffer_length_; /**< buffer size for internal use */
  Check_t check_type_;   /**< check type for data set */
  std::vector params_; /**< configuration of data reader sparse input */
  std::shared_ptr checker_; /**< checker aim to perform error check of the input data */
  bool skip_read_{false};            /**< set to true when you want to stop the data reading */
  int current_record_index_{0};
  size_t total_slot_num_;
  std::vector last_batch_nnz_;

  Tensor2 temp_host_dense_buffer_;  // read data to make checker move
  Tensor2 host_dense_buffer_;
  std::vector> host_sparse_buffer_;
}

其构建代码如下，需要注意，

• 有一个继承于基类的变量 std::shared_ptr buffer_ 指向的是 ThreadBuffer。
• 变量 host_sparse_buffer_ 是构建在 Host 之上，而非GPU之上，这个 host_sparse_buffer_ 作用是文件中读取数据，解析成csr，放置到 host_sparse_buffer_ 之上。
• 关于变量 DataReaderSparseParam 的说明，这是一个DataReaderSparseParam 数组，如果做如下设置，则 params_ 包含三个元素，分别对应分了 user, good, cate。

model.add(hugectr.Input(label_dim = 1, label_name = "label",
                        dense_dim = 0, dense_name = "dense",
                        data_reader_sparse_param_array =
                        [hugectr.DataReaderSparseParam("UserID", 1, True, 1),
                        hugectr.DataReaderSparseParam("GoodID", 1, True, 11),
                        hugectr.DataReaderSparseParam("CateID", 1, True, 11)]))

DataReaderWorker 具体定义如下：

DataReaderWorker(const int worker_id, const int worker_num,
                 const std::shared_ptr& gpu_resource, int* loop_flag,
                 const std::shared_ptr& buffer, const std::string& file_list,
                 size_t buffer_length, bool repeat, Check_t check_type,
                 const std::vector& params)
    : IDataReaderWorker(worker_id, worker_num, gpu_resource, !repeat, loop_flag, buffer),
      buffer_length_(buffer_length),
      check_type_(check_type),
      params_(params),
      total_slot_num_(0),
      last_batch_nnz_(params.size(), 0) {

  total_slot_num_ = 0;
  for (auto& p : params) {
    total_slot_num_ += p.slot_num;
  }
  source_ = std::make_shared(worker_id, worker_num, file_list, repeat);
  create_checker();

  int batch_size = buffer->batch_size;
  int batch_size_start_idx = buffer->batch_size_start_idx;
  int batch_size_end_idx = buffer->batch_size_end_idx;
  int label_dim = buffer->label_dim;
  int dense_dim = buffer->dense_dim;

  CudaCPUDeviceContext ctx(gpu_resource->get_device_id()); // 得到了本worker对应哪个GPU
  std::shared_ptr> buff =
      GeneralBuffer2::create();

  buff->reserve({static_cast(batch_size_end_idx - batch_size_start_idx),
                 static_cast(label_dim + dense_dim)},
                &host_dense_buffer_);
  buff->reserve({static_cast(label_dim + dense_dim)}, &temp_host_dense_buffer_);

  for (auto& param : params) {
    host_sparse_buffer_.emplace_back(batch_size * param.slot_num,
                                     batch_size * param.max_feature_num);
  }

  buff->allocate();
}

具体拓展如下，其中每个thread里面含有一个worker：

或者我们进一步简化几个内存类，得到如下，DataReaderWorker 操作 DataReader 之中的一个 ThreadBuffer，

4.7 读取数据

Reader构建时候，会建立一个 checker_，用来从文件读取数据。

4.7.1 Checker

void create_checker() {
  switch (check_type_) {
    case Check_t::Sum:
      checker_ = std::make_shared(*source_);
      break;
    case Check_t::None:
      checker_ = std::make_shared(*source_);
      break;
    default:
      assert(!"Error: no such Check_t && should never get here!!");
  }
}

以 CheckNone 为例，可以看到其就是读取文件。

class CheckNone : public Checker {
 private:
  const int MAX_TRY{10};

 public:
  CheckNone(Source& src) : Checker(src) {}
  /**
   * Read "bytes_to_read" byte to the memory associated to ptr.
   * Users don't need to manualy maintain the check bit offset, just specify
   * number of bytes you really want to see in ptr.
   * @param ptr pointer to user located buffer
   * @param bytes_to_read bytes to read
   * @return `DataCheckError` `OutOfBound` `Success` `UnspecificError`
   */
  Error_t read(char* ptr, size_t bytes_to_read) noexcept {
    try {
      Checker::src_.read(ptr, bytes_to_read);
      return Error_t::Success;
    } catch (const std::runtime_error& rt_err) {
      std::cerr << rt_err.what() << std::endl;
      return Error_t::BrokenFile;
    }
  }

  /**
   * Start a new file to read.
   * @return `FileCannotOpen` or `UnspecificError`
   */
  Error_t next_source() {
    for (int i = MAX_TRY; i > 0; i--) {
      Error_t flag_eof = Checker::src_.next_source();
      if (flag_eof == Error_t::Success || flag_eof == Error_t::EndOfFile) {
        return flag_eof;
      }
    }
    CK_THROW_(Error_t::FileCannotOpen, "Checker::src_.next_source() == Error_t::Success failed");
    return Error_t::FileCannotOpen;  // to elimate compile error
  }
};

4.7.2 CSR 样例

我们从 samples/ncf/preprocess-1m.py 之中找出一个代码来看看 csr 文件的格式。

def write_hugeCTR_data(huge_ctr_data, filename='huge_ctr_data.dat'):
    with open(filename, 'wb') as f:
        #write header
        f.write(ll(0)) # 0: no error check; 1: check_num
        f.write(ll(huge_ctr_data.shape[0])) # the number of samples in this data file
        f.write(ll(1)) # dimension of label
        f.write(ll(1)) # dimension of dense feature
        f.write(ll(2)) # long long slot_num
        for _ in range(3): f.write(ll(0)) # reserved for future use

        for i in tqdm.tqdm(range(huge_ctr_data.shape[0])):
            f.write(c_float(huge_ctr_data[i,2])) # float label[label_dim];
            f.write(c_float(0)) # dummy dense feature
            f.write(c_int(1)) # slot 1 nnz: user ID
            f.write(c_uint(huge_ctr_data[i,0]))
            f.write(c_int(1)) # slot 2 nnz: item ID
            f.write(c_uint(huge_ctr_data[i,1]))

4.7.3 读取批次数据

read_a_batch 完成具体解析数据集工作。

• 首先从文件读取数据。
• 等待 ThreadBuffer（就是DataReader的thread_buffers_成员变量）的状态变成ReadyForWrite。
• 解析成csr，放入到 host_dense_buffer_。
• 调用 wait_until_h2d_ready 等待拷贝完成。
• 其次调用cudaMemcpyAsync把数据从 host_dense_buffer_ 拷贝到 ThreadBuffer 之中。这里有两点很重要：
  • 目前数据在 host_sparse_buffer_（CPU）之上，需要拷贝到 GPU（目标是 ThreadBuffer 的 device_sparse_buffers 成员变量）。
  • 而且，host_sparse_buffer_ 是 CSR 格式，ThreadBuffer 的 device_sparse_buffers 成员变量是SparseTensor 格式，需要转换。
  • 这里是通过拷贝就进行了转换。

有几点如下：

• nnz 的意思是：non-zero feature number。
• 每一个slot数据对应了一个CSR row。

具体代码如下：

  /**
   * read a batch of data from data set to heap.
   */
  void read_a_batch() {
    // 得到各种配置
    long long current_batch_size = buffer_->batch_size;
    int label_dim = buffer_->label_dim;
    int dense_dim = buffer_->dense_dim;
    int label_dense_dim = label_dim + dense_dim;
    int batch_size_start_idx = buffer_->batch_size_start_idx;
    int batch_size_end_idx = buffer_->batch_size_end_idx;

    try {
      if (!checker_->is_open()) {
        read_new_file(); // 读一个新文件
      }
    } catch (const internal_runtime_error& rt_err) {
      Error_t err = rt_err.get_error();
      if (err == Error_t::EndOfFile) { // 文件读完了
        if (!wait_until_h2d_ready()) return;  // 等待 buffer_ 状态变为 ReadyForWrite
        buffer_->current_batch_size = 0;
        assert(buffer_->state.load() == BufferState::Writing); // 设置
        is_eof_ = true;
        buffer_->state.store(BufferState::ReadyForRead); // 设置状态为可读

        while (buffer_->state.load() != BufferState::ReadyForWrite) {
          usleep(2);
          if (*loop_flag_ == 0) return;  // in case main thread exit
        }
        return;  // need this return to run from begining
      } else {
        throw;
      }
    }

    // if the EOF is faced, the current batch size can be changed later
    
    for (auto& each_csr : host_sparse_buffer_) {
      each_csr.reset();
    }
    // batch loop
    for (int batch_idx = 0; batch_idx < buffer_->batch_size; ++batch_idx) {//读取batch中一个
      if (batch_idx >= current_batch_size) { // 如果已经读取batch之中的全部数据了
        for (size_t param_id = 0; param_id < params_.size(); ++param_id) { // 多个embedding
          // 如果是前面那个例子，这里遍历的就是user, good, cate
          auto& param = params_[param_id];
          // host_sparse_buffer_类型是std::vector>
          auto& current_csr = host_sparse_buffer_[param_id]; 
          for (int k = 0; k < param.slot_num; k++) { // slot数目就是行数
            current_csr.new_row(); // 增加一行
          }
        }
        if (batch_idx >= batch_size_start_idx &&
            batch_idx < batch_size_end_idx) {  // only read local device dense data
          // 设置dense
          float* ptr =
              host_dense_buffer_.get_ptr() + (batch_idx - batch_size_start_idx) * label_dense_dim;

          for (int j = 0; j < label_dense_dim; j++) {
            ptr[j] = 0.f;
          }
        }
        continue;
      }
      try {
        try {
          if (batch_idx >= batch_size_start_idx &&
              batch_idx < batch_size_end_idx) {  // only read local device dense data
            // 读取dense参数
            CK_THROW_(checker_->read(reinterpret_cast(host_dense_buffer_.get_ptr() +
                                                             (batch_idx - batch_size_start_idx) *
                                                                 label_dense_dim),
                                     sizeof(float) * label_dense_dim),
                      "failure in reading label_dense");
          } else {
            // 读取dense参数
            CK_THROW_(checker_->read(reinterpret_cast(temp_host_dense_buffer_.get_ptr()),
                                     sizeof(float) * label_dense_dim),
                      "failure in reading label_dense");
          }

          for (size_t param_id = 0; param_id < params_.size(); ++param_id) {
            auto& current_csr = host_sparse_buffer_[param_id];
            current_csr.set_check_point();
          }
          // 读取sparse参数
          for (size_t param_id = 0; param_id < params_.size(); ++param_id) {
            auto& param = params_[param_id];
            auto& current_csr = host_sparse_buffer_[param_id];
            for (int k = 0; k < param.slot_num; k++) {
              int nnz; // 读取一个int到nnz，就是得到nnz的大小，non-zero feature number
              CK_THROW_(checker_->read(reinterpret_cast(&nnz), sizeof(int)),
                        "failure in reading nnz");
              current_csr.new_row(); // 换行
              size_t num_value = current_csr.get_num_values();
              // 读取nnz个数据
              CK_THROW_(checker_->read(reinterpret_cast(
                                           current_csr.get_value_tensor().get_ptr() + num_value),
                                       sizeof(T) * nnz),
                        "failure in reading feature_ids_");
              current_csr.update_value_size(nnz);
            }
          }
        } catch (const internal_runtime_error& rt_err) { // 回退
          batch_idx--;  // restart i-th sample
          for (auto& each_csr : host_sparse_buffer_) {
            each_csr.roll_back();
          }
          Error_t err = rt_err.get_error();
          if (err == Error_t::DataCheckError) {
            ERROR_MESSAGE_("Error_t::DataCheckError");
          } else {            // Error_t::BrokenFile, Error_t::UnspecificEror, ...
            read_new_file();  // can throw Error_t::EOF
          }
        }

        current_record_index_++;

        // start a new file when finish one file read
        if (current_record_index_ >= data_set_header_.number_of_records) {
          read_new_file();  // can throw Error_t::EOF
        }
      } catch (const internal_runtime_error& rt_err) {
        Error_t err = rt_err.get_error();
        if (err == Error_t::EndOfFile) {
          current_batch_size = batch_idx + 1;
        } else {
          throw;
        }
      }
    }

    for (auto& each_csr : host_sparse_buffer_) {
      each_csr.new_row();
    }
    
    // do h2d
    // wait buffer and schedule
		// 目前数据在 host_sparse_buffer_（CPU）之上，需要拷贝到 GPU（目标是 ThreadBuffer 的 device_sparse_buffers 成员变量），使用 cudaMemcpyHostToDevice
    // 而且，host_sparse_buffer_ 是 CSR 格式，ThreadBuffer 的 device_sparse_buffers 成员变量是SparseTensor格式，需要转换
    if (!wait_until_h2d_ready()) return;
    buffer_->current_batch_size = current_batch_size;
    {
      CudaCPUDeviceContext context(gpu_resource_->get_device_id());
      // 目标是 ThreadBuffer 的 device_sparse_buffers 成员变量
      auto dst_dense_tensor = Tensor2::stretch_from(buffer_->device_dense_buffers);
      CK_CUDA_THROW_(cudaMemcpyAsync(dst_dense_tensor.get_ptr(), host_dense_buffer_.get_ptr(),
                                     host_dense_buffer_.get_size_in_bytes(), cudaMemcpyHostToDevice,
                                     gpu_resource_->get_memcpy_stream()));

      for (size_t param_id = 0; param_id < params_.size(); ++param_id) { // 遍历嵌入层
        auto dst_sparse_tensor =
            SparseTensor::stretch_from(buffer_->device_sparse_buffers[param_id]);
        if (buffer_->is_fixed_length[param_id] &&
            last_batch_nnz_[param_id] == host_sparse_buffer_[param_id].get_num_values()) {
          // 拷贝到GPU，同时也进行了转换，提取了CSR的成员变量，拷贝到了SparseTensor的对应地址
          CK_CUDA_THROW_(cudaMemcpyAsync(dst_sparse_tensor.get_value_ptr(),
                                         host_sparse_buffer_[param_id].get_value_tensor().get_ptr(),
                                         host_sparse_buffer_[param_id].get_num_values() * sizeof(T),
                                         cudaMemcpyHostToDevice,
                                         gpu_resource_->get_memcpy_stream()));
        } else {
          // 拷贝到GPU
          sparse_tensor_helper::cuda::copy_async(dst_sparse_tensor, host_sparse_buffer_[param_id],
                                                 gpu_resource_->get_memcpy_stream());
          last_batch_nnz_[param_id] = host_sparse_buffer_[param_id].get_num_values();
        }
      }
      // 进行同步
      CK_CUDA_THROW_(cudaStreamSynchronize(gpu_resource_->get_memcpy_stream()));
    }
    assert(buffer_->state.load() == BufferState::Writing);
    buffer_->state.store(BufferState::ReadyForRead);
  }
};

4.7.3.1 等待

这里wait_until_h2d_ready会等待。

bool wait_until_h2d_ready() {
  BufferState expected = BufferState::ReadyForWrite;
  while (!buffer_->state.compare_exchange_weak(expected, BufferState::Writing)) {
    expected = BufferState::ReadyForWrite;
    usleep(2);
    if (*loop_flag_ == 0) return false;  // in case main thread exit
  }
  return true;
}

4.7.3.2 读取文件

read_new_file 完成了对文件的读取。

void read_new_file() {
  constexpr int MAX_TRY = 10;
  for (int i = 0; i < MAX_TRY; i++) {
    if (checker_->next_source() == Error_t::EndOfFile) {
      throw internal_runtime_error(Error_t::EndOfFile, "EndOfFile");
    }

    Error_t err =
        checker_->read(reinterpret_cast(&data_set_header_), sizeof(DataSetHeader));
    current_record_index_ = 0;
    if (!(data_set_header_.error_check == 0 && check_type_ == Check_t::None) &&
        !(data_set_header_.error_check == 1 && check_type_ == Check_t::Sum)) {
      ERROR_MESSAGE_("DataHeaderError");
      continue;
    }
    if (static_cast(data_set_header_.slot_num) != total_slot_num_) {
      ERROR_MESSAGE_("DataHeaderError");
      continue;
    }
    if (err == Error_t::Success) {
      return;
    }
  }
  CK_THROW_(Error_t::BrokenFile, "failed to read a file");
}

4.7.4 小结

我们总结逻辑如下，线程一直调用 data_reader_thread_func_ 来循环读取：

另外一个逻辑视角是：

1. 多线程调用 data_reader_thread_func_，其使用 read_a_batch 从数据文件之中读取数据解析为CSR。每一个embedding层 对应一个CSR。
2. CSR 被放入 DataReaderWorker 的 host_sparse_buffer_。
3. 随着batch不断读取，CSR 行数在不断增加，每一个slot对应了一行，所以一个batch的行数就是 batch_size * slot_num。
4. 使用 cudaMemcpyAsync 把CSR从 host_sparse_buffer_ 拷贝到ThreadBuffer（位于GPU）。ThreadBuffer是 SparseTensor 类型了。
5. 目前CSR数据就在 GPU 之上了。

这里简化了多GPU，多worker 的情况。

0x05 读取到embedding

我们接下来看看 DataCollector，就是流水线的第二级，就是这里的黄色框 "Copy to GPU"。其实其内部文字修改为：Copy To Embedding 更合适。

5.1 DataCollector

我们首先看看DataCollector的定义，这里省略了成员函数，主要成员变量是。

• std::shared_ptr broadcast_buffer_ : CPU 数据拷贝到 GPU 之上，GPU 上就在这里。
• std::shared_ptr output_buffer_ ：这 个就是 DataReaderOutput，就是 Reader 的成员变量，复制到这里是为了 collector 操作方便。
• BackgroundDataCollectorThread background_collector_ ：线程主体，主要包括 ThreadBuffer 和 BroadcastBuffer，会把数据从 ThreadBuffer 拷贝到 BroadcastBuffer 之上。
• std::thread background_collector_thread_ ：工作线程。

/**
 * @brief A helper class of data reader.
 *
 * This class implement asynchronized data collecting from heap
 * to output of data reader, thus data collection and training
 * can work in a pipeline.
 */
template 
class DataCollector {
  
  class BackgroundDataCollectorThread {
    std::vector> thread_buffers_;
    std::shared_ptr broadcast_buffer_;

    std::atomic loop_flag_;
    int counter_;
    std::vector last_batch_nnz_;  // local_gpu_count * embedding number
    std::vector worker_status_;
    int eof_worker_num_;

    std::shared_ptr resource_manager_;
  }
  
  std::shared_ptr broadcast_buffer_;
  std::shared_ptr output_buffer_;

  BackgroundDataCollectorThread background_collector_;
  std::thread background_collector_thread_;

  std::atomic loop_flag_;
  std::vector last_batch_nnz_;

  std::shared_ptr resource_manager_;
};

目前具体如下，Collector 之中的 broadcast_buffer_ 和 output_buffer_ 都指向了GPU，但GPU之中尚且没有数据：

5.2 ThreadBuffer 2 BroadBuffer

5.2.1 工作线程

BackgroundDataCollectorThread 的作用是把数据从 DataReader 的thread_buffers_拷贝到 broadcast_buffer_。

class BackgroundDataCollectorThread {
  std::vector> thread_buffers_;
  std::shared_ptr broadcast_buffer_;

  std::atomic loop_flag_;
  int counter_;
  std::vector last_batch_nnz_;  // local_gpu_count * embedding number
  std::vector worker_status_;
  int eof_worker_num_;

  std::shared_ptr resource_manager_;

 public:
  BackgroundDataCollectorThread(const std::vector> &thread_buffers,
                                const std::shared_ptr &broadcast_buffer,
                                const std::shared_ptr &resource_manager)
      : thread_buffers_(thread_buffers),
        broadcast_buffer_(broadcast_buffer),
        loop_flag_{true},
        counter_{0},
        last_batch_nnz_(
            broadcast_buffer->is_fixed_length.size() * resource_manager->get_local_gpu_count(),
            0),
        worker_status_(thread_buffers.size(), 0),
        eof_worker_num_(0),
        resource_manager_(resource_manager) {}
  
  void start() {
    
    while (loop_flag_.load()) {
      // threadbuffer是源数据，broadcast buffer是目标数据
      auto ¤t_src_buffer = thread_buffers_[counter_];
      auto &dst_buffer = broadcast_buffer_;
      auto src_expected = BufferState::ReadyForRead; // 期望源数据是这个状态
      auto dst_expected = BufferState::ReadyForWrite; // 期望目标数据是这个状态

      if (worker_status_[counter_]) {
        counter_ = (counter_ + 1) % thread_buffers_.size();
        continue;
      }

      if ((current_src_buffer->state.load() == BufferState::Reading ||
           current_src_buffer->state.compare_exchange_weak(src_expected, BufferState::Reading)) &&
          (dst_buffer->state.load() == BufferState::Writing ||
           dst_buffer->state.compare_exchange_weak(dst_expected, BufferState::Writing))) {

        // 如果源数据是可读或者正在读，并且，目标数据是可写或者正在写，则可以操作
        
        if (current_src_buffer->current_batch_size == 0) {
          worker_status_[counter_] = 1;
          eof_worker_num_ += 1;
          current_src_buffer->state.store(BufferState::FileEOF);
        }
        if (static_cast(eof_worker_num_) != thread_buffers_.size() &&
            current_src_buffer->current_batch_size == 0) {
          counter_ = (counter_ + 1) % thread_buffers_.size();
          dst_buffer->state.store(BufferState::ReadyForWrite); // 设定目标数据的状态
          continue;
        }
        dst_buffer->current_batch_size = current_src_buffer->current_batch_size;
        if (current_src_buffer->current_batch_size != 0) {
          // 进行广播操作
          broadcast(current_src_buffer, dst_buffer, last_batch_nnz_, resource_manager_);

          current_src_buffer->state.store(BufferState::ReadyForWrite); // 设定目标数据的状态
          counter_ = (counter_ + 1) % thread_buffers_.size();
        } else {
          memset(worker_status_.data(), 0, sizeof(char) * worker_status_.size());
          eof_worker_num_ = 0;
          counter_ = 0;
        }

        dst_buffer->state.store(BufferState::ReadyForRead); // 会通知源数据可以继续读取了
      } else {
        usleep(2); // 否则等待一会
      }
    }
  }

  void stop() { loop_flag_.store(false); }
};

5.2.2 拷贝操作

这里就是从源数据拷贝到目标数据，并且是逐个参数进行拷贝。这个是设备之内的拷贝。

template 
void broadcast(const std::shared_ptr& thread_buffer,
               std::shared_ptr& broadcast_buffer,
               std::vector& last_batch_nnz_,
               const std::shared_ptr& resource_manager) {
  int param_num = thread_buffer->param_num;
  int dense_dim = thread_buffer->dense_dim;
  int label_dim = thread_buffer->label_dim;
  int batch_size = thread_buffer->batch_size;
  int batch_size_per_gpu = batch_size / resource_manager->get_global_gpu_count();
  int local_gpu_count = resource_manager->get_local_gpu_count();

#pragma omp parallel for num_threads(local_gpu_count)
  for (int i = 0; i < local_gpu_count; ++i) { // 遍历本地的GPU
    
    auto local_gpu = resource_manager->get_local_gpu(i);
    CudaDeviceContext ctx(local_gpu->get_device_id());

    for (int param_id = 0; param_id < param_num; ++param_id) { // 遍历嵌入层
      // 从 thread_buffer 拷贝到 broadcast_buffer
      auto src_sparse_tensor =
          SparseTensor::stretch_from(thread_buffer->device_sparse_buffers[param_id]);
      auto dst_sparse_tensor =
          SparseTensor::stretch_from(broadcast_buffer->sparse_buffers[i * param_num + param_id]);

      // 拷贝sparse参数
      if (thread_buffer->is_fixed_length[param_id] &&
          last_batch_nnz_[i * param_num + param_id] == src_sparse_tensor.nnz()) {
        CK_CUDA_THROW_(cudaMemcpyAsync(dst_sparse_tensor.get_value_ptr(),
                                       src_sparse_tensor.get_value_ptr(),
                                       src_sparse_tensor.nnz() * sizeof(T),
                                       cudaMemcpyDeviceToDevice, local_gpu->get_p2p_stream()));
      } else {
        sparse_tensor_helper::cuda::copy_async(dst_sparse_tensor, src_sparse_tensor,
                                               cudaMemcpyDeviceToDevice,
                                               local_gpu->get_p2p_stream());
        last_batch_nnz_[i * param_num + param_id] = src_sparse_tensor.nnz();
      }
    }

    // 拷贝dense参数
    auto dst_dense_tensor = Tensor2::stretch_from(broadcast_buffer->dense_tensors[i]);
    auto src_dense_tensor = Tensor2::stretch_from(thread_buffer->device_dense_buffers);
    CK_CUDA_THROW_(cudaMemcpyAsync(
        dst_dense_tensor.get_ptr(),
        src_dense_tensor.get_ptr() + i * batch_size_per_gpu * (label_dim + dense_dim),
        batch_size_per_gpu * (label_dim + dense_dim) * sizeof(float), cudaMemcpyDeviceToDevice,
        local_gpu->get_p2p_stream()));
    
    // 同步
    CK_CUDA_THROW_(cudaStreamSynchronize(local_gpu->get_p2p_stream()));
  }
}

逻辑如下，多了一步从 ThreadBuffer 到 BroadcastBuffer 的操作。

5.3 读取到output

目前的流程是：DataFile ---> Host buffer ----> ThreadBuffer ----> BroadcastBuffer。

现在数据已经拷贝到了 GPU 之上的 BroadcastBuffer，我们需要看看最后训练时候怎么拿到数据。

5.3.1 Train

我们首先回到 train 函数，其调用了 read_a_batch_to_device_delay_release 来从 BroadcastBuffer 拷贝数据。

bool Session::train() {
  try {
    // 确保 train_data_reader_ 已经启动
    if (train_data_reader_->is_started() == false) {
      CK_THROW_(Error_t::IllegalCall,
                "Start the data reader first before calling Session::train()");
    }

#ifndef DATA_READING_TEST
    // 需要 reader 先读取一个 batchsize 的数据。
    long long current_batchsize = train_data_reader_->read_a_batch_to_device_delay_release(); // 读取数据
    if (!current_batchsize) {
      return false; // 读不到就退出，没有数据了
    }
    #pragma omp parallel num_threads(networks_.size()) //其后语句将被networks_.size()个线程并行执行
    { 
      
      size_t id = omp_get_thread_num();
      CudaCPUDeviceContext ctx(resource_manager_->get_local_gpu(id)->get_device_id());
      cudaStreamSynchronize(resource_manager_->get_local_gpu(id)->get_stream());
    }
    // reader 可以开始解析数据
    train_data_reader_->ready_to_collect();
#ifdef ENABLE_PROFILING
    global_profiler.iter_check();
#endif

    // If true we're gonna use overlaping, if false we use default
    if (solver_config_.use_overlapped_pipeline) {
      train_overlapped();
    } else {
      for (const auto& one_embedding : embeddings_) {
        one_embedding->forward(true); // 嵌入层进行前向传播，即从参数服务器读取embedding，进行处理
      }

      // Network forward / backward
      if (networks_.size() > 1) {
        // 单机多卡或多机多卡
        // execute dense forward and backward with multi-cpu threads
        #pragma omp parallel num_threads(networks_.size())
        {
          // dense网络的前向反向
          size_t id = omp_get_thread_num();
          long long current_batchsize_per_device =
              train_data_reader_->get_current_batchsize_per_device(id);
          networks_[id]->train(current_batchsize_per_device); // 前向操作
          const auto& local_gpu = resource_manager_->get_local_gpu(id);
          local_gpu->set_compute_event_sync(local_gpu->get_stream());
          local_gpu->wait_on_compute_event(local_gpu->get_comp_overlap_stream());
        }
      } else if (resource_manager_->get_global_gpu_count() > 1) {
        // 多机单卡
        long long current_batchsize_per_device =
            train_data_reader_->get_current_batchsize_per_device(0);
        networks_[0]->train(current_batchsize_per_device); // 前向操作
        const auto& local_gpu = resource_manager_->get_local_gpu(0);
        local_gpu->set_compute_event_sync(local_gpu->get_stream());
        local_gpu->wait_on_compute_event(local_gpu->get_comp_overlap_stream());
      } else {
        // 单机单卡
        long long current_batchsize_per_device =
            train_data_reader_->get_current_batchsize_per_device(0);
        networks_[0]->train(current_batchsize_per_device); // 前向操作
        const auto& local_gpu = resource_manager_->get_local_gpu(0);
        local_gpu->set_compute_event_sync(local_gpu->get_stream());
        local_gpu->wait_on_compute_event(local_gpu->get_comp_overlap_stream());
        networks_[0]->update_params();
      }

      // Embedding backward
      for (const auto& one_embedding : embeddings_) {
        one_embedding->backward(); // 嵌入层反向操作
      }

      // Exchange wgrad and update params
      if (networks_.size() > 1) {
        #pragma omp parallel num_threads(networks_.size())
        {
          size_t id = omp_get_thread_num();
          exchange_wgrad(id); // 多卡之间交换dense参数的梯度
          networks_[id]->update_params();
        }
      } else if (resource_manager_->get_global_gpu_count() > 1) {
        exchange_wgrad(0);
        networks_[0]->update_params(); 
      } 
      for (const auto& one_embedding : embeddings_) {
        one_embedding->update_params(); // 嵌入层更新sparse参数
      }

      // Join streams
      if (networks_.size() > 1) {
        #pragma omp parallel num_threads(networks_.size())
        {
          size_t id = omp_get_thread_num();
          const auto& local_gpu = resource_manager_->get_local_gpu(id);
          local_gpu->set_compute2_event_sync(local_gpu->get_comp_overlap_stream());
          local_gpu->wait_on_compute2_event(local_gpu->get_stream());
        }
      }
      else {
        const auto& local_gpu = resource_manager_->get_local_gpu(0);
        local_gpu->set_compute2_event_sync(local_gpu->get_comp_overlap_stream());
        local_gpu->wait_on_compute2_event(local_gpu->get_stream());
      }
      return true;
    }
#else
      data_reader_->read_a_batch_to_device();
#endif

  } catch (const internal_runtime_error& err) {
    std::cerr << err.what() << std::endl;
    throw err;
  } catch (const std::exception& err) {
    std::cerr << err.what() << std::endl;
    throw err;
  }
  return true;
}

5.3.2 read_a_batch_to_device_delay_release

read_a_batch_to_device_delay_release 是最终配置好embedding数据的地方。

long long read_a_batch_to_device_delay_release() override {
  current_batchsize_ = data_collector_->read_a_batch_to_device();
  return current_batchsize_;
}

我们看看 read_a_batch_to_device。这里 read_a_batch_to_device_delay_release 和 read_a_batch_to_device 是沿用旧版本命名，已经和目前状况不符合。

具体逻辑是：看看 broadcast_buffer_ 的状态是不是可以读取 ReadyForRead，如果不可以，就等一会。如果可以，就继续，即遍历GPU，逐个从broadcast拷贝到output（也是设备之间的拷贝），也对 label 和 dense 进行split。

  long long read_a_batch_to_device() {

    BufferState expected = BufferState::ReadyForRead;
    while (!broadcast_buffer_->state.compare_exchange_weak(expected, BufferState::Reading)) {
      expected = BufferState::ReadyForRead;
      usleep(2);
    }
    long long current_batch_size = broadcast_buffer_->current_batch_size;
    if (current_batch_size != 0) {
      int local_gpu_count = resource_manager_->get_local_gpu_count();

#pragma omp parallel for num_threads(local_gpu_count)
      for (int i = 0; i < local_gpu_count; ++i) {
        auto local_gpu = resource_manager_->get_local_gpu(i);
        CudaDeviceContext ctx(local_gpu->get_device_id());

        // wait until last iteration finish
        auto label_tensor = Tensor2::stretch_from(output_buffer_->label_tensors[i]);
        auto label_dense_tensor = Tensor2::stretch_from(broadcast_buffer_->dense_tensors[i]);

        // 遍历 sparse 参数
        for (size_t param_id = 0; param_id < output_buffer_->sparse_name_vec.size(); ++param_id) {
          const auto &top_name = output_buffer_->sparse_name_vec[param_id];
          int idx_broadcast = i * broadcast_buffer_->param_num + param_id;
          // broadcast 的是源
          auto src_sparse_tensor =
              SparseTensor::stretch_from(broadcast_buffer_->sparse_buffers[idx_broadcast]);
          if (output_buffer_->sparse_tensors_map.find(top_name) ==
              output_buffer_->sparse_tensors_map.end()) {
            CK_THROW_(Error_t::IllegalCall, "can not find sparse name");
          }
          // output是目标
          auto dst_sparse_tensor =
              SparseTensor::stretch_from(output_buffer_->sparse_tensors_map[top_name][i]);

          // 从broadcast拷贝到output
          if (broadcast_buffer_->is_fixed_length[idx_broadcast] &&
              last_batch_nnz_[idx_broadcast] == src_sparse_tensor.nnz()) {
            CK_CUDA_THROW_(cudaMemcpyAsync(dst_sparse_tensor.get_value_ptr(),
                                           src_sparse_tensor.get_value_ptr(),
                                           src_sparse_tensor.nnz() * sizeof(T),
                                           cudaMemcpyDeviceToDevice, local_gpu->get_stream()));
          } else {
            // 从broadcast拷贝到output
            sparse_tensor_helper::cuda::copy_async(dst_sparse_tensor, src_sparse_tensor,
                                                   cudaMemcpyDeviceToDevice,
                                                   local_gpu->get_stream());
            last_batch_nnz_[idx_broadcast] = src_sparse_tensor.nnz();
          }
        }
        const int label_dense_dim = output_buffer_->label_dense_dim;

        // 拷贝label和dense
        if (output_buffer_->use_mixed_precision) {
          auto dense_tensor = Tensor2<__half>::stretch_from(output_buffer_->dense_tensors[i]);
          // 进行分块
          split(label_tensor, dense_tensor, label_dense_tensor, label_dense_dim,
                local_gpu->get_stream());
        } else {
          auto dense_tensor = Tensor2::stretch_from(output_buffer_->dense_tensors[i]);
          split(label_tensor, dense_tensor, label_dense_tensor, label_dense_dim,
                local_gpu->get_stream());
        }
      }
    } else {
      broadcast_buffer_->state.store(BufferState::ReadyForWrite);
    }
    return current_batch_size;
  }

5.3.3 split

label 和 dense 早已经拷贝到了GPU之上，这步做的是分成block，然后使用 GPU thread 进行操作。

template 
__global__ void split_kernel__(int batchsize, float* label_ptr, int label_dim, TypeComp* dense_ptr,
                               int dense_dim, const float* label_dense, int label_dense_dim) {
  int idx = blockDim.x * blockIdx.x + threadIdx.x;
  if (idx < batchsize * label_dense_dim) {
    const int in_col = idx % label_dense_dim;
    const int in_row = idx / label_dense_dim;
    const int out_row = in_row;
    if (in_col < label_dim) {
      const int out_col = in_col;
      label_ptr[out_row * label_dim + out_col] = label_dense[idx];
    } else {
      const int out_col = in_col - label_dim;
      dense_ptr[out_row * dense_dim + out_col] = label_dense[idx];
    }
  }
  return;
}

template 
void split(Tensor2& label_tensor, Tensor2& dense_tensor,
           const Tensor2& label_dense_buffer, const int label_dense_dim,
           cudaStream_t stream) {
  // check the input size
  assert(label_tensor.get_dimensions()[0] == dense_tensor.get_dimensions()[0]);
  assert(label_tensor.get_num_elements() + dense_tensor.get_num_elements() ==
         label_dense_buffer.get_num_elements());

  const int batchsize = label_tensor.get_dimensions()[0];
  const int label_dim = label_tensor.get_dimensions()[1];
  const int dense_dim = dense_tensor.get_dimensions()[1];
  const int BLOCK_DIM = 256;
  const int GRID_DIM = (label_dense_buffer.get_num_elements() - 1) / BLOCK_DIM + 1;

  if (dense_dim > 0) {
    split_kernel__<<>>(
        batchsize, label_tensor.get_ptr(), label_dim, dense_tensor.get_ptr(), dense_dim,
        label_dense_buffer.get_ptr(), label_dense_dim);
  } else if (dense_dim == 0) {
    split_kernel__<<>>(
        batchsize, label_tensor.get_ptr(), label_dim, (TypeComp*)0, 0, label_dense_buffer.get_ptr(),
        label_dense_dim);

  } else {
    CK_THROW_(Error_t::WrongInput, "dense_dim < 0");
  }

  return;
}

这样后续就可以训练了，后续是通过 finalize_batch 之中进行读取。

void finalize_batch() {
  for (size_t i = 0; i < resource_manager_->get_local_gpu_count(); i++) {
    const auto &local_gpu = resource_manager_->get_local_gpu(i);
    CudaDeviceContext context(local_gpu->get_device_id());
    CK_CUDA_THROW_(cudaStreamSynchronize(local_gpu->get_stream()));
  }

  broadcast_buffer_->state.store(BufferState::ReadyForWrite);
}

template 
void AsyncReader::ready_to_collect() {
  auto raw_device_id = reader_impl_->get_last_batch_device();
  auto local_gpu = resource_manager_->get_local_gpu(raw_device_id);
  CudaDeviceContext ctx(local_gpu->get_device_id());
  CK_CUDA_THROW_(cudaEventRecord(completion_events_[raw_device_id], local_gpu->get_stream()));

  reader_impl_->finalize_batch(&completion_events_[raw_device_id]);
}

0x06 总结

具体逻辑如下，本章节之中，各个buffer之间拷贝，是依据其状态是 ReadyForRead 和 ReadyForWrite 来完成的。最终sparse 参数的embedding是在DataReaderOutput，即后续 GPU 上的计算是从output开始的。

0xFF 参考

https://developer.nvidia.com/blog/introducing-merlin-hugectr-training-framework-dedicated-to-recommender-systems/

https://developer.nvidia.com/blog/announcing-nvidia-merlin-application-framework-for-deep-recommender-systems/

https://developer.nvidia.com/blog/accelerating-recommender-systems-training-with-nvidia-merlin-open-beta/

HugeCTR源码阅读

embedding层如何反向传播

https://web.eecs.umich.edu/~justincj/teaching/eecs442/notes/linear-backprop.html

稀疏矩阵存储格式总结+存储效率对比:COO,CSR,DIA,ELL,HYB