1

我用“-np 2”调用 mpirun。我将等级 0 的进程称为“主”，将等级 1 的进程称为“从”。

目标：

  1. master 偶尔会向 slave 发送消息，例如mpi::send(1, UPDATE, data);. 其他消息类型包括 DIE、COMPUTE ...等。这些消息类型是具有唯一值的常量整数。
  2. slave 运行一个无限循环，“监听”来自 master 的任何消息。当它收到一条消息时，它会向主设备发送一个确认。

执行：

从运行：

...
int updateData, computeData;
mpi::request updateRequest = world.irecv(0,UPDATE, updateData);    
mpi::request computeRequest = world.irecv(0,COMPUTE, computeData);    

do {
  cerr << "slave ready to take a command" << endl;
  if(updateRequest.test()) {
    cerr << "slave ireceived UPDATE" << endl;
    world.send(0, UPDATE_ACK, 0);
    cerr << "slave sent UPDATE_ACK" << endl;

    /* do something useful 
    ...
    ...
    */

    updateRequest = world.irecv(0, UPDATE, updateData);

  } else if (computeRequest.test()) {
    ...
  } else {
    boost::this_thread::sleep( boost::posix_time::seconds(1) );
  }
}

当主人运行时：

...
world.send(1, UPDATE, 10);
cerr << "master sent UPDATE" << endl;
int dummy;
world.recv(1, UPDATE_ACK, dummy);
cerr << "master received UPDATE_ACK" << endl;
...

主代码的更多上下文：

...
// update1
world.send(1, UPDATE, params);
cerr << "master sent UPDATE" << endl;
int dummy;
world.recv(1, UPDATE_ACK, dummy);
cerr << "master received UPDATE_ACK" << endl;

// update2
world.send(1, UPDATE2, params2);
cerr << "master sent UPDATE2" << endl;
world.recv(1, UPDATE2_ACK, dummy);
cerr << "master received UPDATE2_ACK" << endl;

// update3
world.send(1, UPDATE3, params3);
cerr << "master sent UPDATE3" << endl;
world.recv(1, UPDATE3_ACK, dummy);
cerr << "master received UPDATE3_ACK" << endl;

...

// training iterations
do {
  
  mpi::request computeRecvReq1, computeRecvReq2;
  std::map<int, int> result1, result2;

  // for each line in a text file, the master asks the slave(s)
  // to compute two things and aggregates the results
  for(unsigned sentId = 0; sentId != data.size(); sentId++) {

    // these two functions won't return until at least one slave is "idle"
    CollectSlavesWork1(computeRecvReq1, result1);
    CollectSlavesWork2(computeRecvReq2, result2);

    // async ask the slave to compute and async get the results
    world.isend(1, COMPUTE, sentId);
    computeRecvReq1 = world.irecv(1, RESULT1, result1);
    computeRecvReq2 = world.irecv(1, RESULT2, result2);

  }

  // based on the slave(s) work, the master updates params1 
  // and send them again to the slave(s)
  world.send(1, UPDATE, params);
  cerr << "master sent UPDATE" << endl;
  world.recv(1, UPDATE_ACK, dummy);              // PROBLEM HAPPENS HERE
  cerr << "master received UPDATE_ACK" << endl;


} while(!ModelIsConverged())

...

输出：

...

奴隶准备接受命令

主发送更新

奴隶收到更新

从机发送 UPDATE_ACK

主控收到 UPDATE_ACK

奴隶准备接受命令

...

奴隶准备接受命令

主发送更新

奴隶收到更新

从机发送 UPDATE_ACK

奴隶准备接受命令

...

问题： 主人第一次发送更新消息时，一切似乎都很好。但是，在第二次中，主服务器没有收到 UPDATE_ACK。

4

0 回答 0