tensorflow v0.9中目前在用的devcie assignment算法是simple placer算法，相比于白皮书中cost model算法实现简单。simpler placer算法

优先选择/gpu:0设备， 但

不支持 multi gpu assignment。

白皮书提到的cost model可以根据设备资源代价、数据传输代价平衡分配设备，在v0.9版本中有部分实现，但还未开放使用，见 core/graph/costmodel.cc 

 

simple_placer的实现代码在文件python/core/common_runtime/simple_placer.cc，其中包含device_assignment的核心功能。

core/common_runtime/simple_placer_test.cc测试片段如下

1 //// 2 // 3 // A SimplePlacerTest method has three phases: 4 // 5 // 1. Build a TensorFlow graph, with no (or partial) device assignments. 6 // 2. Attempt to compute a placement using the SimplePlacer. 7 // 3. EITHER: test that the constraints implied by the graph are respected; 8 //    or that an appropriate error was reported. 9 //10 ////11 class SimplePlacerTest : public ::testing::Test {12  protected:13   SimplePlacerTest() {14     // Build a set of 10 GPU and 10 CPU devices.15     // NOTE: this->local_devices_ owns the device objects;16     // this->devices_ contains borrowed pointers to the device17     // objects.18     for (int i = 0; i < 10; ++i) {    // 添加了10 cpu和10 gpu的fake devices19       local_devices_.emplace_back(FakeDevice::MakeCPU(20           strings::StrCat("/job:a/replica:0/task:0/cpu:", i)));21       devices_.AddDevice(local_devices_.back().get());22       // Insert the GPUs in reverse order.23       local_devices_.emplace_back(FakeDevice::MakeGPU(24           strings::StrCat("/job:a/replica:0/task:0/gpu:", 9 - i)));25       devices_.AddDevice(local_devices_.back().get());26     }27   }28   ...29 }30 ...31 // Test that a graph with no constraints will successfully assign nodes to the32 // "best available" device (i.e. prefer GPU over CPU).33 TEST_F(SimplePlacerTest, TestNoConstraints) {34   Graph g(OpRegistry::Global());35   {  // Scope for temporary variables used to construct g.   // 用GraphDefBuilder构建graph的结构36     GraphDefBuilder b(GraphDefBuilder::kFailImmediately);37     Node* input = ops::SourceOp("TestInput", b.opts().WithName("in"));    38     ops::UnaryOp("TestRelu", ops::NodeOut(input, 0), b.opts().WithName("n1"));39     ops::UnaryOp("TestRelu", ops::NodeOut(input, 1), b.opts().WithName("n2"));40     TF_EXPECT_OK(BuildGraph(b, &g));   //  BuildGraph函数将GraphDefBuilder的图写入到Graph中41   }42  43   TF_EXPECT_OK(Place(&g));   // Place函数将graph中的node布放到设备列表中44   EXPECT_DEVICE_TYPE(g, "in", DEVICE_CPU);   // 期望：input节点在CPU中，n1节点在GPU中，n2节点在GPU中，故而GPU优先级大于CPU45   EXPECT_DEVICE_TYPE(g, "n1", DEVICE_GPU);46   EXPECT_DEVICE_TYPE(g, "n2", DEVICE_GPU);47 }

其中BuildGraph函数将GraphDefBuilder 对象中的graph 结构定义写入到Graph中。Place函数将graph中的node布放到设备列表中，其中device assignment算法的核心在SimplePlacer::Run函数中

1  // Builds the given graph, and (if successful) indexes the node 2   // names for use in placement, and later lookup. 3   Status BuildGraph(const GraphDefBuilder& builder, Graph* out_graph) { 4     TF_RETURN_IF_ERROR(builder.ToGraph(out_graph)); 5     nodes_by_name_.clear(); 6     for (Node* node : out_graph->nodes()) { 7       nodes_by_name_[node->name()] = node->id(); 8     } 9     return Status::OK();10   }11   // Invokes the SimplePlacer on "graph". If no DeviceSet is specified, the12   // placement will use the default DeviceSet (of 10 CPU and 10 GPU devices).13   //14   // REQUIRES: "*graph" was produced by the most recent call to BuildGraph.15   Status Place(Graph* graph, DeviceSet* devices, SessionOptions* options) {16     SimplePlacer placer(graph, devices, options);17     return placer.Run();18   }

SimplePlacer::Run()在core/common_runtime/simple_placer.cc文件中，具体实现分为4个步骤： 

步骤1和2： 遍历graph的node，将node加入到ColocationGraph对象中（不包含source和sink节点）。

1 // 1. First add all of the nodes. Note that steps (1) and (2) 2 // requires two passes over the nodes because the graph (and hence 3 // the constraints) may not be acyclic.  这里graph可能是有环的？ 4 for (Node* node : graph_->nodes()) { 5     // Skip the source and sink nodes. 6     if (!node->IsOp()) { continue; } 7     status = colocation_graph.AddNode(*node); 8     if (!status.ok()) return AttachDef(status, node->def()); 9   }10 // 2. Enumerate the constraint edges, and use them to update the disjoint node set.         // disjoint set（并查集，即不相交的节点集合），一种树型数据结构，11 ...

1 ColocationGraph maintains the connected components of a colocation constraint graph, and uses this information to assign a satisfying device placement to the nodes of the graph. 2 The implementation uses the union- find algorithm to maintain the connected components efficiently and incrementally as edges (implied by ColocationGraph::ColocateNodes() invocations) are added.  3 参考：

 

 

步骤3：如下图和code所示，source和sink节点分配在cpu上，已指定device的节点不再重新分配。分配方式有方面，见Heuristic A和

Heuristic

 B。

1  3. For each node, assign a device based on the constraints in thedisjoint node set. 2   std::vector
      
        devices; 3   std::vector
       
         second_pass; 4   for (Node* node : graph_->nodes()) { 5     // Skip the source and sink nodes. 6     if (!node->IsOp()) { 7       continue; 8     } 9     // Skip nodes that already have an assigned name.10     if (!node->assigned_device_name().empty()) {11       continue;12     }13     // Heuristic A: prefer to place "generators" with their only14     // consumers.15     //16     // If this is a node with no inputs and a single (non-ref)17     // consumer, we save this for a second pass, so that the18     // consumer's placement is chosen.19     if (IsGeneratorNode(node)) {    // generator node: no input, one output, not a reference-type node20       second_pass.push_back(node);21       continue;22     }23     status = colocation_graph.GetDevicesForNode(node, &devices);24     ...25     // Returns the first device in sorted devices list so we will always26     // choose the same device.27     //28     // TODO(vrv): Factor this assignment out into a pluggable29     // algorithm, so that SimplePlacer is responsible for enforcing30     // preconditions and we can experiment with other algorithms when31     // given a choice of devices. Once we have a better idea of the32     // types of heuristics we want to use and the information needed33     // to perform good placement we can add an interface for this.34     string assigned_device = devices[0]->name();35     // Heuristic B: If the node only operates on metadata, not data,36     // then it is desirable to place that metadata node with its37     // input.38     if (IsMetadataNode(node)) {   39       // Make sure that the input device type is in the list of supported40       // device types for this node.41       const Node* input = (*node->in_edges().begin())->src();42       // TODO(vrv): if the input is empty, consider postponing this43       // node's assignment to the second pass, so that we handle the44       // case where a metadata node's input comes from a backedge45       // of a loop.46       const string& input_device_name = input->assigned_device_name();47       if (CanAssignToDevice(input_device_name, devices)) {48         assigned_device = input_device_name;49       }50     }51     AssignAndLog(assigned_device, node);   // 将assigned_device分配个node节点，在步骤3中没有对符合Heuristic A的GeneratorNode分配设备，而是在步骤4中完成的52   }
       
      

1 bool IsGeneratorNode(const Node* node) {2   return node->num_inputs() == 0 && node->num_outputs() == 1 && node->out_edges().size() == 1 && !IsRefType(node->output_type(0));3 }

1 bool IsMetadataNode(const Node* node) {2   const string& node_type = node->type_string();3   return (node_type == "Size" || node_type == "Shape" || node_type == "Rank");4 }

步骤4：给步骤3中的Generator Node分配device。

// 4. Perform a second pass assignment for those nodes explicitly skipped during the first pass....

 

部分参考：

  （韩文的-_-）

” “