使用 boost::asio 提升线程池

我正在尝试使用带有工作队列的boost asio创建提升线程池。但是我卡在某一点上，我需要一个监视功能，它应该继续运行并跟踪队列。我不确定我该怎么写，目前我把它放在我的线程池类中。我也在尝试编写具有全局函数的分离线程。有人可以提出任何建议吗？

#include <boost/thread/thread.hpp>
#include <boost/asio.hpp>
#include<memory>
#include<vector>
#include<list>
using namespace std;
class threadWork
{
public:
virtual void run()=0;
};
class thread_pool 
{ 
private: 
  boost::asio::io_service io_service_; 
  boost::asio::io_service::work work_; 
  boost::thread_group threads_; 
  std::size_t available_; 
  boost::mutex mutex_;
  list<shared_ptr<threadWork>> workQueue;
public: 

  thread_pool( std::size_t pool_size ) : work_( io_service_ ), available_( pool_size ) 
  { 
    for ( std::size_t i = 0; i < pool_size; ++i ) 
    { 
      threads_.create_thread( boost::bind( &boost::asio::io_service::run, &io_service_ ) ); 
    } 
  } 

  ~thread_pool() 
  { 
    io_service_.stop(); 
try 
    { 
      threads_.join_all(); 
    } 
    catch ( ... ) {} 
  } 
  void enqueue(shared_ptr<threadWork> work)
  {
  workQueue.push_back(work);
  }
  void keepRunning() // how to call this ?
  {
  while(true)
  {
      boost::unique_lock< boost::mutex > lock( mutex_ ); 
      if ( 0 == available_ ) // If no threads are available, then sleep. 
      {
          boost::this_thread::sleep(boost::posix_time::milliseconds(1000));
      }
      else
      {
          if(workQueue.empty() != true)
          {
               --available_;  
               io_service_.post( boost::bin(&thread_pool::wrap_task,this , workQueue));
               workQueue.pop_front();
          }
      }
  }
  }
 private: 
  void wrap_task(list<shared_ptr<threadWork>>& workQueue ) 
  { 
    try 
    { 
    workQueue.front()->run(); // Run the user supplied task. 
    } 
catch ( ... )  // Suppress all exceptions. 
{
} 
boost::unique_lock< boost::mutex > lock( mutex_ );  
    ++available_; 
  } 
}; 
class someWork:public threadWork
{
public:
virtual void run()
{
    cout<<"some long task n";
     boost::this_thread::sleep(boost::posix_time::milliseconds(5000));
}
};
int main()
{
thread_pool pool(10);
pool.keepRunning(); // this stuck code so where to start this ?
shared_ptr<threadWork> w(new someWork);
pool.enqueue(w);

return 0;
}